A survey of United States dermatologists’ knowledge, attitudes, and practices with intramuscular triamcinolone

Since the 1970s, intramuscular triamcinolone (IMT) has been available as an option for systemic corticosteroid use in dermatology. Although shown to be safe and effective in early studies, this method of systemic corticosteroid delivery fell out of favor in the 1980s in many United States residency programs. To identify factors associated with US dermatologists’ preferences for and use of IMT we surveyed a random sample of US board-certified dermatologists to assess knowledge, attitudes, and practices regarding IMT in dermatologists’ daily clinical practice. A total of 844 out of 2000 dermatologists completed the survey (42.2%). Only 55.0% reported feeling comfortable using IMT for steroid-responsive dermatoses, while 90.4% felt comfortable using oral corticosteroids for steroid-responsive dermatoses. Most participants (59.2%) did not prefer IMT over oral corticosteroids when both were indicated. One third (33.3%) of the participants reported that none of the faculty during their residency advocated using IMT. Receiving education on IMT indications (OR = 1.96 [95% CI: 1.46–2.63]) and encouragement to use IMT (OR = 4.29 [95% CI: 3.01–6.11]) during residency were positively associated with use of IMT at least once a month in current practice. Current knowledge, attitudes, and practices surrounding IMT vary amongst practicing dermatologists. Modifiable factors such as training could improve comfort with use of this short-term systemic steroid treatment modality.


Introduction
Since their discovery, corticosteroids have been widely used as an effective therapy to treat numerous inflammatory and autoimmune disorders. However, corticosteroids are associated with serious risks, especially when used long-term and/ or at higher doses [1][2][3][4]. Intramuscular triamcinolone (IMT) has been available since the 1970s as a systemic corticosteroid option for a large array of steroid-responsive dermatoses [5,6]. Safety and effectiveness of this treatment were demonstrated, and use of this modality was widespread until a decline in the 1980s, around the time that publications advocated against the use of IMT because of inability to conduct alternate-day therapy or to immediately discontinue therapy, and theoretical increased hypothalamic-pituitary-adrenal axis suppression compared to oral therapy [7][8][9][10][11]. More recently, IMT was again demonstrated to be a safe and effective treatment, and a comprehensive review suggested that IMT can be regarded favorable to other glucocorticoid modalities in determined clinical situations [5,7]. One study evaluated IMT use in treating various steroid-responsive dermatologic conditions and showed that adrenocorticotropic hormone levels were unaffected while significant improvements in the conditions were observed by patients and physicians [5]. No significant clinical side effects were observed [5]. Based on these findings, Reddy et al. concluded that IMT appears to be a safe and effective treatment option for steroid-responsive dermatologic diseases, and the impact on the adrenocortical system should not discourage its use [5].
Side effects of IMT include lipodystrophy, acne, dysmenorrhea, sterile abscesses, avascular necrosis and easy bruising, but only lipodystrophy and sterile abscesses are specific to intramuscular delivery and are rare with proper technique [19][20][21][22][23][24]. When compared to oral therapy, this modality also offers greater adherence, arguably a more favorable side effect profile, and more precise dosing due to the elimination of patient error such as skipping days or incorrect dosing [25]. It has been suggested that IMT could, therefore, reasonably become first-line therapy for many steroid-responsive dermatoses requiring systemic therapy [3,25,26]. Despite this, many US dermatology residents may receive little or no training on the use of IMT, though comfort with IMT use increases with frequency of use and previous training on its use [13].
We therefore conducted a survey of board-certified US dermatologists to assess their knowledge, attitudes, and practices (KAP) toward IMT, and their experiences with IMT during training. Our primary objective was to quantify how many currently practicing US dermatologists used IMT in the last month and their comfort level associated with this treatment option. Secondary objectives included surveying dermatologists' preferred uses for IMT and the related degree of training they received during residency.

Methods
We conducted an anonymous cross-sectional survey study of KAP of IMT use in a random sample of 2000 US boardcertified dermatologists supplied by the American Academy of Dermatology in 2017. This study received ethical approval from the University of Utah Institutional Review Board (#75532).

Survey development
The survey consisted of 29 questions: eight about training, five about comfort with prescribing corticosteroids, seven about practice patterns, and nine about demographics (see online supplement 1). Skip patterns were used for certain survey questions. Questions were developed by a group of board-certified dermatologists in the Department of Dermatology at the University of Utah. Face validity and comprehensibility were tested by surveying 30 dermatologists and dermatology residents at the University of Utah, which narrowed the survey to the final 29 questions described here.

Study population and sampling
We requested contact information (mailing address) from the American Academy of Dermatology for a random sample of 2000 board-certified dermatologists in the US and associated territories (Puerto Rico, District of Columbia, and US military bases) from a larger pool of all active US dermatologist members of the AAD in the database. Based on prior surveys conducted by the AAD, we expected a response rate of 25% thus 2000 respondents were estimated to be needed to identify adequate representation. Invitations to participate in the survey were sent via mail up to three times with a self-addressed stamped return envelope. In the cover letter, a tinyURL and QR code were also provided for participants to complete the same survey online via REDCap.

Statistical analyses
This survey employed a single-stage, simple random sample without replacement design. Descriptive statistics included frequencies (%). IMT usage was divided into at least once a month and less than once a month, and odds of using IMT at least monthly was calculated using binary logistic regression. Statistical analyses were conducted using SPSS v25 (IBM, Armonk, New York) and Stata v16.1 statistical software (StataCorp, College Station, Texas).

Results
A total of 2000 individuals were invited to participate in this study (see online supplement 2), and of these, 844 returned the survey (response rate = 42.2%). Participant demographics are shown in Table 1. Four hundred ninety-four of these participants (58.5%) were male, and 540 (64.0%) were ≥ 50 years old. Respondents practiced in 49 of 50 of states; Washington, DC; Puerto Rico; and US military bases. Four hundred fourteen participants (49.1%) reported practicing in a suburban setting, and 52.0% (439/844) provided for a community of ≥ 250,000 people. Seven hundred ninety-five (94.2%) were at least five years out from residency, and 45.1% (381/844) had been practicing for ≥ 25 years. Seven hundred thirty-five respondents (87.1%) marked that general dermatology was their primary specialty, and 81% (686/844) worked in private practice. The strong majority of private practice dermatologists (86.4%, 593/686) saw ≥ 75 patients/week, while less than two thirds of academic dermatologists (64.7%, 86/133) saw ≥ 75 patients/ week.

Training on intramuscular triamcinolone
Dermatology residency types included academic, military, private, and combined university and private. Most (55.2%, 466/844) reported a class size of 3 to 5 residents per year (Table 1). During residency, one-third reported that none of the faculty advocated using IMT, one-third indicated that few faculty members advocated its use, and one-third reported that at least half the faculty did promote the treatment (Table 3). Regarding usage, 59.5% (502/844) reported having been trained on IMT dosing and 65.2% (550/844) on IMT indications. However, 56.4% (476/844) disagreed when asked if they were encouraged to use IMT for steroid-responsive dermatoses, and 67.9% (573/844) disagreed when asked if they had been encouraged to use this over oral corticosteroids for the same indication ( Table 3).

Comparison of practice characteristics with usage
Bivariate logistic regression models were used to analyze the relationship between practice demographics and use of IMT at least once a month (Table 4). Dermatologists aged ≥ 65 years were more likely to use IMT at least monthly compared to those younger than 50 (OR = 1.48 [95% CI: 1.04-2.11]), and dermatologists with ≥ 35 years of practice were more likely to use IMT at least monthly

Comparison of residency training with usage
Bivariate logistic regression models were used to analyze the relationship between residency training variables and use of IMT at least monthly (   . Those who agreed that they were encouraged to use IMT for steroid-responsive dermatoses during residency had 4.29 (95% CI: 3.01-6.11, n = 843) fold increased odds of using IMT monthly compared to those who were neutral or disagreed about being encouraged to do so. Those who agreed they were encouraged to use IMT over oral corticosteroids for steroid-responsive dermatoses had 3.32 (95% CI: 2.06-5.33) fold increased odds of using monthly IMT compared to those who reported that they were neutral or disagreed about being encouraged to do so. Dermatologists who agreed that they were comfortable using IMT had 70.82 (95% CI: 36.47-137.53) fold increased odds of using IMT monthly compared to those who disagreed that they were comfortable using IMT.

Discussion
This survey study gives insight into dermatologists' KAP and their relationship to training on the use of IMT in clinical practice. IMT usage among dermatologists varied considerably between never users versus use > 1 time/day. While most were comfortable using IMT for steroid-responsive dermatoses, IMT was not favored by most participants when both oral corticosteroids and IMT were indicated. IMT was preferred over oral corticosteroids most commonly when patients requested this modality or when the dermatologist desired increased patient adherence. Notably, physician anticipation of better patient satisfaction than oral therapy was also less frequently noted. Together, these data seem to suggest a sense of reluctance from clinicians to use this modality except in certain situations, which is supported by previous literature [6,7,13,27]. Nearly half of dermatologists in our study reported using IMT to the same extent as they did during residency. Residency training often forms the foundation for future practice [33][34][35]. Most dermatologists in our study felt they were not encouraged to use IMT for steroid-responsive dermatoses during training. Conversely, we did observe that encouragement to use IMT during residency increased the likelihood of regularly administering IMT (≥ 1 time/month) in practice. These findings highlight the association between residency training and encouragement to use a certain modality on future practice.
Older, more experienced, and busier dermatologists were more likely to administer IMT at least monthly. This is possibly due to how common IMT use was during the 1980s.
Increased billing cost and health insurance reimbursement associated with IMT when compared to oral corticosteroids [7], could help explain why dermatologists in private practice, who tend to see more patients [36], were more likely to regularly use IMT than academic dermatologists. The previously cited need for prospective, double-blind, placebocontrolled studies with larger numbers of subjects to confirm the efficacy of IMT in different dermatological condition and the lack of direct comparisons between oral and IMT modalities are also possible deterrents for IMT use. [5] Limitations Given the survey-based design of the study, recall bias could impact results. In particular, perception of the degree that IMT was encouraged as a resident, a question that represents a construct from many years prior for some dermatologists, could be skewed by participants' current practice and memory. Many in the sample also completed residency several decades ago and, thus, their training experiences might not reflect current training practices. Lastly, though our response rate was better than the AAD's own membership-derived surveys, [37][38][39] the response rate could bias the results if those who responded were fundamentally different than those who did not, thus impacting the representativeness of the study sample.

Conclusions
These findings suggest that many dermatologists do not commonly use IMT nor are comfortable using it for steroidresponsive dermatoses even when such method is clinically indicated and could offer specific benefits. The strong associations between current practice habits and residency training regarding IMT use highlight the importance of residency training, a modifiable factor. Residency programs should consider including training on safe and efficient methods of systemic corticosteroid delivery such as IMT in the appropriate clinical setting to expand the available treatment options that can be offered and increase comfort in the use of such, consequently providing patients with more choices and potentially improving patient satisfaction. Further, dermatologists should be trained and educated on IMT so that as continuing evidence is found, dermatologists will be comfortable and knowledgeable of proper IMT use.
Author contributions G.K. wrote the main manuscript text and A.S. prepared online supplements and tables. All authors reviewed the manuscript.
Funding No external funding was obtained specifically for this study. The study was funded by the University of Utah Department of Dermatology. The use of REDCap for the research reported in this publication was supported by the National Center for Advancing Translational Sciences of the National Institutes of Health under Award Number 8UL1TR000105 (formerly UL1RR025764). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.