Practice patterns of physical therapists and physical therapist assistants treating patients with breast cancer related lymphedema

Clinical Practice Guidelines (CPGs) aim to improve patient outcomes through implementation of proven interventions and decrease variation in practices. The relevance of this article is to describe the current state practice of physical therapists who diagnose and treat patients with Breast Cancer Related Lymphedema (BCRL). It also provides a description of physical therapist-reported adherence to the BCRL CPG recommendations which establishes the need for implementation interventions to improve adherence. The purpose of this study is to describe practice patterns of physical therapists (PT) and physical therapist assistants (PTA) who treat patients with breast cancer-related lymphedema and determine if they are adherent to best evidence recommendations for lymphedema diagnosis and intervention. An electronic survey to collect practice pattern data of PTs and PTAs who treat patients with BCRL was distributed. A descriptive and quantitative statistical analysis was performed. Twenty-six percent of respondents read the American Physical Therapy Association sponsored lymphedema diagnosis clinical practice guideline (CPG) and 20% read the lymphedema intervention CPG. Lymphoscintigraphy was the only diagnosis or intervention tool with a significant difference in use between therapists who read versus did not read the CPGs. Adherence to “should do” recommendations was variable: bioimpedance (18.2%), volume calculation (49.3%), ultrasound (0%), patient reported outcome tools (64.9%), compression garments (43.9%), exercise (87.2%), and compression bandaging (56.8%). There is variability in adherence to recommendations for both the lymphedema diagnosis and intervention CPGs. Interventions to improve implementation and adherence to CPG recommendations are warranted.


Background
Breast cancer-related lymphedema (BCRL) is a progressive, chronic, incurable, and potentially debilitating diagnosis that is linked to decreased quality of life, increased functional limitations, and, in severe cases, disability [1]. Significant variation exists in the physical therapy diagnosis and treatment of BCRL [2][3][4][5]. Over the past several years, two clinical practice guidelines (CPGs) have been published to provide guidance to physical therapists in these areas [2,3]. According to a 1992 report from the Institute of Medicine, CPGs are "systematically developed statements to assist practitioner and patient decisions about appropriate health care for specific clinical circumstances", (p. 8) [6]. CPGs are intended to improve patient outcomes through implementation of proven interventions, decrease variation in practices, reduce treatment disparities, empower patients, and influence public policy [7,8]. CPGs are based on a systematic review of the highest quality literature by a multidisciplinary panel of experts. While it is crucial that clinicians utilize high-quality research in their decision-making process, it is also important to consider their clinical experience and the values and preferences of the patient [9].
Sackett et al. [10] defined evidence-based medicine (EBM) as "the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients" (p. 71) and later redefined it as "the integration of best available evidence, clinical expertise and patient preferences and values" [11] (p. 1). EBM is supported by the development and implementation of CPGs [7]. One goal of CPG development is to increase the implementation of evidence into clinical practice; however there is variable evidence to determine if this being achieved [12][13][14].
Physical therapy management of patients with BCRL should incorporate an evidence-based approach to maximize patient outcomes. To promote the best practice of physical therapy diagnosis and treatment of BCRL, two CPGs were developed by the American Physical Therapy Association's (APTA) Academy of Oncologic Physical Therapy with companion papers outlining implementation strategies [2,3,[15][16][17][18]. Both CPGs were written by Guideline Development Groups who critically appraised the literature to determine the level of evidence and to grade the clinical practice recommendations [19][20][21][22]. Recommendation statements were assigned a letter grade which corresponds to a descriptor of the strength of the recommendation (strong, moderate, weak, theoretical/foundational, expert opinion, or best practice). The Intervention CPG methods included linking level of evidence and grade of recommendation via language of obligation, such as "must or should," "should," "may," or "should or may" [22].
Over the past several years, many CPGs have been developed to guide physical therapy diagnosis and intervention with variable levels of clinician adherence documented in the literature [13,[23][24][25]. To date, there is no published description of current physical therapist practice patterns regarding the APTA sponsored lymphedema diagnosis and intervention CPGs. The purpose of this study was to describe the practice patterns of physical therapists (PT) and physical therapist assistants (PTA) who treat patients with BCRL and determine if they are practicing within the recommendations of the APTA's lymphedema diagnosis and intervention CPGs.

Sample and design
A cross-sectional web-based survey was conducted using a secure Research Electronic Data Capture (REDCap) service, hosted at Midwestern University-Downers Grove. All aspects of this study were approved by the Midwestern University -Downers Grove Internal Review Board.
The survey was sent to PTs and PTAs who were members of the APTA's Academy of Oncologic Physical Therapy (n=1300) and the Lymphology Association of North America (LANA) (n=2000) in fall of 2021. Participants were asked to share the request to complete the survey with other PTs and PTAs who treat patients with BCRL.

Survey development
The survey was divided into three main sections: participant demographics, barriers and facilitators to implementation of the CPGs, and self-reported practice patterns specific to BCRL. The survey can be found in Online Resource 1. The findings of the barriers and facilitators section of the survey will be reported elsewhere. The questions related to self-reported practice patterns required respondents to indicate frequency of use of diagnostic and intervention tools by BCRL stage. Therapists were asked, "for each stage of lymphedema, please indicate which breast cancer-related diagnostic tools you use most frequently in your clinical setting" and "for each stage of lymphedema, please indicate which interventions you generally recommend to your patients." The diagnostic and intervention tools were selected based on the recommendations from the APTA's lymphedema diagnosis and intervention CPGs [2,3].
The survey was sent to thirteen clinicians with expertise in managing patients with BCRL. Ten reviewed the survey and provided feedback, which was incorporated in the final draft. See Online Resource 1 for the demographic and practice pattern sections of the survey.

Survey distribution
An email was sent to all members of the APTA's Academy of Oncologic Physical Therapy and LANA which contained an invitation to participate in the survey and an electronic link to the survey. Access to the survey was granted after selecting "yes" to the question indicating informed consent and a professional designation of "physical therapist" or "physical therapist assistant." Survey responses were collected over two months between October and December 2021. Electronic reminders were sent two additional times during this period to facilitate a higher response rate.

Statistical analysis
Data analysis was completed using R version 4.1.1 (R Core Team 2021) [26]. Demographic and practice patterns data were analyzed using descriptive statistics and reported as frequencies and percentages. A two-sample proportion z-test was performed for respondents that used the diagnostic or intervention tool to determine if there was a difference in the percentages of respondents who read the CPG and those who did not read the CPG. Significance for this analysis was set a priori at α ≤ 0.05. Adherence to CPG recommendations was calculated utilizing all survey respondents to determine the percentage and 95% confidence interval of those who reported congruence with "should do" CPG recommendations [27].

Results
The results of this survey were part of a larger study evaluating the barriers and facilitators to implementing the BCRL diagnosis and intervention CPGs. The data on barriers and facilitators to implementation will be reported elsewhere. A total of 231 surveys were returned with 180 responses to questions related to clinical practice patterns. There were a small number of non-responders to most practice pattern questions which affected frequency percentages reported in the manuscript.

Demographics
A total of 180 participants consented to take the practice patterns portion of the survey which included acknowledging their profession as PT or PTA. Of those participants, 145 (81%) were PTs, 3 (2%) were PTAs, and 32 (17%) did not designate a professional designation. Over two-thirds (n=113, 76%) of the respondents had over 11 years of experience as a PT or PTA and were predominantly female (n=142, 96%). Nearly half were members of the APTA's Academy of Oncologic Physical Therapy while only 18 (12%) held their American Board of Physical Therapy Specialties (ABPTS) Oncologic Specialization. The majority, 137 (93%), were Certified Lymphedema Therapists (CLT) and 103 (70%) were LANA Certified. Details of reported demographics can be found in Table 1.

Practice patterns
One hundred forty-four respondents completed the questions about the frequency of use of BCRL diagnostic tools of which 38 (26%) read the diagnostic CPG and 106 (74%) did not. Membership to the APTA was associated (p<0.001) with having read both the diagnosis and intervention CPGs, while LANA certification (p<0.001) was associated with reading the diagnosis CPG. No other characteristics were associated with reading the CPGs. Diagnostic tool frequency use is reported as a percentage of use for those who read the CPG (RCPG) and those who did not read the CPG (NRCPG) in Fig. 1 and reflects use of the tool at any stage of BCRL. The most commonly used diagnostic tools for both groups were clinical examination 38 (100%) RCPG, 100 (94%) NRCPG; circumference measures using 2 centimeters difference 25 (66%) RCPG, 78 (74%) NRCPG; circumference The only diagnostic measure that had a statistically significant difference in the proportion of use between those who read versus those who did not read the CPG was lymphoscintigraphy 7 (18%) RCPG, 4 (4%) NRCPG, (p-value = 0.01).
One hundred forty-seven PTs and PTAs completed questions about the frequency of use of BCRL intervention techniques of which 29 (20%) had read the intervention CPG and 118 (80%) did not read. Intervention tool frequency use is reported in Fig. 2 and reflects use of the tool at any stage of BCRL. No statistically significant differences in percentages of intervention-related practice pattern usage between those who read and those who did not read were found (p-value range 0.126-0.99). Acupuncture 3 (10%) RCPG, 10 (9%) NRCPG and low-level laser therapy 7 (24%) RCPG, 21 (18%) NRCPG were the least frequently used; while compression garments 29

Adherence with clinical practice guidelines
Adherence rate to "should do" recommendations from the diagnosis and intervention CPGs are reported in Table 2. All survey respondents, regardless of whether they read or did not read the CPGs were included in this analysis (n=144). Survey respondents had to report they used the tool at the appropriate stage of BCRL as recommended in

Discussion
This is the first study to our knowledge that describes the practice patterns of PTs and PTAs in relationship to the APTA's lymphedema diagnosis and intervention CPGs. This description provides insight into the PT and PTAs' familiarity with the CPGs and with what frequency they are using the diagnostic and intervention tools with patients who have BCRL. Only 26% of respondents read the diagnostic CPG and 20% read the intervention CPG. There was no statistically significant difference in those having read either CPG between therapists with a doctoral degree or APTA specialty certification than those who did not. However, therapists who were members of the APTA were more likely to read both CPGs and members of LANA were more likely to read the diagnosis CPG. The findings of this current study are lower than reported for other physical therapy CPGs. Corkery et al. (2014) reported 77% of PTs surveyed were familiar with CPGs for whiplash, but only 39% of them used the CPG more than 75% of the time. Willy et al. [28] reported 48.8% of PTs read the patellofemoral pain CPG. While there were relatively few respondents who had read the APTA's lymphedema diagnosis or intervention CPGs, there were no significant differences in percentages of diagnostic or intervention practice patterns except for the use of lymphoscintigraphy. This may be related to the consistency of training of best practices between lymphedema certification programs.
The most commonly utilized diagnostic tools were clinical examination, circumference measures using 2 centimeters difference, circumference measures using volume calculation, and patient reported outcome   [2]. Compression garments, exercise, education, manual lymphatic drainage, and myofascial techniques were the most commonly used intervention techniques. Whether clinicians read or did not read the intervention CPG, there were adherent in using exercise (87.2%) and compression bandaging (56.8%) at the appropriate stage of lymphedema. The variability in adherence to intervention-based CPGs is not unique to the lymphedema CPG. In a recent systematic review, Zadro et al. [13] identified the median percentage of PTs who chose recommended practices for musculoskeletal impairments was 54% with adherence ranging anywhere from 35% for low back pain to 93% for knee and hip arthroplasty and shoulder pain. De Souza et al. [30] found that PTs demonstrated full adherence to back pain CPGs across 6 patient cases 5-24% of the time and partial adherence 32-72% of the time highlighting the need for wider dissemination of CPGs. Adherence to recommendations for stroke physical therapy have been reported between 5-71% [31][32][33]. The barriers to implementing physical therapy CPGs are varied and not well described, likely contributing to the variability in adherence.

Strengths and limitations
The strengths of our study include the large distribution of our survey and the intentional design of our survey questions to discern PT and PTA adherence to lymphedema diagnosis and intervention CPG recommended practice patterns. While our survey response rate was not as high as we had hoped, we were able to distribute the survey to the two largest professional organizations for PTs and PTAs who treat the diagnosis of lymphedema (LANA and Academy of Oncologic Physical Therapy of the APTA). There are a couple of limitations to our study. The first is our inability to calculate a response rate for our survey. We allowed for snowball sampling and utilized the LANA membership list to send out our survey. LANA is a multidisciplinary organization, and the survey was sent to the entire membership list. Respondents did have to attest that they were PTs or PTAs to continue with the survey. These two sampling methods limited our ability to account for the total number of PTs or PTAs who received the survey, thereby limiting our ability to calculate a survey response rate. Other physical therapy practice pattern surveys report a 2-82% response rate [24,28,29,34]. The second is that we assessed adherence to clinical practice guidelines using a single method, survey. Disparities can exist between self-reported data and actual performance in adherence to CPGs [13,35]. Surveys have other weaknesses, including failure to evaluate the clinical context of the environment and patient characteristics which may influence the therapist's choice to adhere to CPG recommendations.

Future research
Interventions to improve implementation and adherence to CPG recommendations are warranted [14]. Implementation interventions should be explicit to be reproducible, yet flexible to be generalizable to multiple clinic and health system settings. Researchers need to strive to assess effectiveness with valid methods and consider multi-faceted assessment of clinician adherence as well as assessment of improved patient outcomes when CPGs are implemented.