A medical chart audit to assess endocrinologist perceptions of the burden of endogenous Cushing’s syndrome

This quantitative, cross-sectional study was conducted to collect disease burden data pertaining to patients with CS from qualified physician respondents. All study materials were reviewed and granted exemption by a central Institutional Review Board (IRB) prior to study execution (Advarra; Columbia, MD; https://​www.​advarra.​com/​). HCPs were recruited via a physician panel through an independent recruitment partner (Toluna) and received an appropriate honorarium for their time participating in the study.

This study was fielded between May 26 and July 27, 2021, and involved the abstraction of retrospective medical chart data from patients with a confirmed diagnosis of CS by endocrinologists. The survey included a 45–60-min web-enabled questionnaire, including a case report form (CRF) component, to capture patient medical chart data and health care practitioner (HCP)-level data in order to assess perceptions of CS disease burden, treatments, and attributes associated with treatments. Considering the rarity of CS, each HCP was required to abstract information from a minimum of 2 patient charts, and a maximum of 8 patient charts.

HCPs were able to participate in the study if they:

Each HCP who qualified to participate provided information via chart abstraction from the medical records of 2–8 patients with CS. The selected medical charts were from patients ≥ 21 years of age who had received a physician confirmed diagnosis of CS at least 3 months before the time of the study, and had received at least one therapy (surgical, radiological, or pharmacological) to treat their CS within the past 12 months. Patients who were diagnosed with adrenal or pituitary carcinomas were excluded.

The data analysis was conducted in SAS 9.4 (SAS Institute Inc., Cary, NC, USA) and Q Research Software 5.6. (Q Research Software, New York, NY). After pilot interviews and throughout the fielding, quality control checks of all the case report forms were conducted to ensure that charts with logical inconsistencies were removed from the sample. Descriptive statistics (such as means, medians, and frequencies) were used to describe the study population across various patient and physician level metrics.

Endocrinologists’ demographic information and practice characteristics are presented in Table 1. A total of 69 endocrinologists were surveyed and they provided information on 273 unique patient charts. The majority of the 69 endocrinologists surveyed (53/69, 73%) were male. The mean (± SD) time in practice was 17.3 (± 7.6) years. The majority of endocrinologists (35/69, 51%) worked in urban practices and were in private practice settings (47/69, 68%) (Table 1). The sample was almost equally distributed between physicians from the northern (26%), southern (29%), eastern (25%) and western (22%) regions of the United States. The mean (± SD) estimated number of patients with endogenous CS seen in the last 6 months was 30 (± 34.4) patients.

^aEndocrinologist were allowed to select multiple practice settings, if applicable

Patient demographics and clinical characteristics at the time of the survey are shown in Table 2. The majority of patients (165/273, 60%) were female with a mean (± SD) age at diagnosis of 40.2 (± 12.3) years and a mean (± SD) age at the most recent visit of 46.5 (± 13.4) years. Mean (± SD) BMI was 33.3 (± 8.3) kg/m², with 50.5% of patients categorized as obese, 33.0% of patients categorized as overweight, 14.7% of patients categorized as normal or healthy weight, and 1.8% of patients categorized as underweight (Table 2). Most patients (167/273, 61%) had private or commercial health insurance. Patient demographics and clinical characteristics at disease diagnosis are shown in Table 2. A majority of patients (194/273, 79%) originally saw their primary care physician (PCP) prior to diagnosis and were diagnosed in a private practice setting (182/273, 67%). At the time of diagnosis, 46/273 patients (17%) had poor health, 107/273 patients (39%) had fair health, 68/273 patients (25%) had neutral health, 45/273 patients (16%) had good health, and 7/273 patients (3%) had excellent health, according to the responding physician.

The patient treatment experience at the time of the study is presented in Table 2. Of the 273 patients, 79 (28.9%) underwent surgery only, 11 patients (4.0%) underwent surgery and radiation therapy, 4 patients (1.4%) underwent radiation therapy and pharmacotherapy, 5 patients (1.8%) underwent surgery, radiation therapy, and pharmacotherapy, 85 patients (31.1%) underwent surgery and pharmacotherapy, 2 patients (< 1%) underwent radiation alone and 87 patients (31.9%) underwent pharmacotherapy alone.

At diagnosis, 34% of patients presented with 1–3 symptoms, 33% of patients presented with 4–6 symptoms, 20% of patients presented with 7–9 symptoms, 8% of patients presented with 10–12 symptoms, and 5% of patients presented with > 13 symptoms (Fig. 1). Symptoms of CS at the time of diagnosis are shown in Fig. 2. The top 10 most common symptoms of CS at the time of diagnosis (Fig. 3) included fatigue, weight gain (in the midsection and upper back), acne, muscle weakness, facial weight gain (i.e., facial roundness), decreased libido, headache, edema, emotional lability, and hirsutism. Although symptoms decreased post-treatment, a large proportion of subjects still exhibited these symptoms post-treatment (Fig. 3). The most commonly reported comorbidities observed in patients with CS at the time of CS diagnosis (i.e., those affecting ≥ 20% of patients) included obesity, hypertension, depression, diabetes, dyslipidemia, anxiety, and impaired glucose tolerance (Table 2).

Healthcare resource utilization was assessed (Table 3). Patients required a mean (± SD) of 1 (± 1.4) hospitalization annually with a mean (± SD) length of impatient stay of 4.3 (± 3.1) days. Patients required a mean (± SD) of 0.6 (± 1.3) annual emergency room (ER) visits, and 4.3 (± 6.3) outpatient visits.

Endocrinologists were asked if they agreed with a series of statements regarding their perception of CS burden and impact on a scale of 1–9, where 1 = Not at all agree and 9 = Completely agree (Fig. 4). The highest proportion of endocrinologists responded “Completely agree” with the statements “CS patients can have reduced ability to function at work or school due to their condition” (percent of endocrinologists who responded “Completely agree” = 35%), “patients with CS feel the impact of their condition every day” (30%), that “CS is a debilitating condition” (28%), “patients with CS often have impaired health-related quality of life” (28%), and “CS results in sleep disturbances that adversely impact patient’s HRQoL” (26%).

Endocrinologists were asked for their perceptions of the most important treatment attributes on a scale of 1 to 5, where 1 = the least important and 5 = the most important (Table 4). The two most important treatment attributes included treatments that were efficacious post-surgery (mean score = 4.0) and efficacious as a combination therapy (3.7). Endocrinologists were asked to rank satisfaction with currently available treatments for CS including surgical intervention, pharmacotherapy, and radiological or other interventions on a scale of 1–9, where 1 = Not at all satisfied and 9 = Extremely satisfied (Table 5). Overall, endocrinologists reported highest satisfaction with surgical intervention with regards to initial efficacy (mean score = 7.2), durability (6.9), safety (6.3), side effects (6.2), tolerability (6.4), and patient’s overall experience (6.9). Endocrinologists also ranked pharmacotherapy higher than radiation therapy for the treatment of CS for initial efficacy (5.9 versus 5.2), safety (5.9 versus 5.4), side effects (5.3 versus 5.2), tolerability (5.7 versus 5.5), and patient’s overall experience (5.9 versus 5.4).

Key factors for evaluating and selecting a CS treatment were rated on a scale of 1–9, with 1 = Not at all important and 9 = Extremely important (Fig. 5). Improving HRQoL (mean score = 7.8) was rated as the most important attribute. Similarly, improving cardiovascular complications/events (e.g., myocardial infarction, stroke, embolism) (7.6), psychiatric symptoms (e.g., depression, anxiety, mood changes) (7.6), skeletal/muscular symptoms (e.g., muscular weakness, decrease in bone mineral density, bone fractures) (7.5), and neurologic symptoms (e.g., headaches, memory, and cognitive difficulties including brain fog) (7.5) were ranked as key factors when choosing CS treatment. While factors in the survey such as “causes high rate of adrenal insufficiency” and “label contains a warning against use in CS” were ranked as less important, none of the factors listed were considered unimportant by physician respondents for choosing CS treatment.

Endocrinologists were asked if they agreed with a series of statements regarding CS treatment and intervention attitudes on a scale of 1–9, where 1 = strongly disagree and 9 = strongly agree (Table 6). The three highest scoring statements were “there is a significant clinical unmet need for patients with endogenous CS” (mean score = 6.6), “better patient support services for CS medications often leads to better patient adherence” (6.5), and “patient out of pocket cost is a significant burden for CS patients on a pharmacological therapy” (6.5). The lowest scoring statement was “patient out of pocket cost is not a significant factor when prescribing pharmacological therapy for my CS patients” (4.6).

During the time in which this study was conducted, additional CS treatments could have been approved, potentially changing the treatment landscape, and thereby altering the proportion of patients that continued to have symptoms after treatment (Fig. 3) or the proportion of patients with a particular comorbidity after treatment. Physician response may have been subject to recall bias; although this may have been mitigated by the use of patient chart data the possibility that details were omitted at the time of patient visits exists. Additionally, when physicians were asked about working in a Center of Excellence, the term was not explicitly defined which may have led to varying interpretations by respondents. Due to the nature of the method used (i.e., a survey given to endocrinologists treating patients at the present time), we have limited historical chart data on the entire medical journey of each patient and all important medical events may not have been captured. For example, treatments administered to patients prior to this study (i.e., those administered by previous doctors or from a different hospital) may not be present in the patients’ charts and were not captured by our survey. Additionally, we did not capture biochemical data to make definitive statements on disease status based on patient cortisol levels. Updated guidelines on cortisol levels indicative of disease severity have recently been issued by the Pituitary Society [18], and a shift toward standardized clinical guidelines may help physicians provide timely and appropriate treatment for patients with CS. Future patient-centered research in CS should focus on identifying biomarkers associated with persistent symptoms after initial treatment, which could influence the development of guidelines for managing ongoing symptoms as current treatments are focused on cortisol management. The cohort of patients with CS included in our study is also not representative of the full spectrum of patients with CS as they were required to have received at least one pharmacological therapy to be eligible for the study. This requirement was added to our eligibility criteria as the aim of our study was to evaluate the burden of illness faced by patients with Cushing’s Syndrome, post-treatment, in the real world. Future studies evaluating concordance between patient chart data and physician perceptions of CS symptoms are also likely to be of interest. Finally, patient symptoms in this study could potentially have been masked due to the use of over-the-counter medications or other prescription treatments not fully captured in charts.