The Influence of Physician and Patient Gender on Risk Assessment for Lung Cancer Resection

Women with large intact abdominal aortic aneurysms remain untreated

OBJECTIVE

A lower elective repair rate among women with abdominal aortic aneurysms (AAAs) has been a consistent finding. Reasons behind this gender gap have not been thoroughly outlined.

METHODS

This was a retrospective multicenter cohort study (ClinicalTrials.gov: NCT05346289) at three European vascular centers in Sweden, Austria and Norway. Patients in surveillance with AAAs were consecutively identified starting from January 1, 2014, until reaching a total sample size of 200 women and 200 men. All individuals were followed for 7 years through medical records. Final treatment distributions and the proportion of "truly untreated" (surgically untreated despite reaching guideline-directed thresholds: 50 mm for women and 55 mm for men) were determined. In a complementary analysis, a universal 55-mm threshold was used. Gender-specific primary reasons behind untreated statuses were clarified. Eligibility for endovascular repair among the truly untreated was assessed in a structured computed tomography analysis.

RESULTS

Women and men had similar median diameters at inclusion (46 mm; P = .54) and at treatment decisions (55 mm; P = .36). After 7 years, the repair rate was lower among women (47% vs 57%). More women were truly untreated (26% vs 8%; P < .001) despite similar mean ages as for male counterparts (79.3 years; P = .16). With the 55-mm threshold, 16% women still classified as truly untreated. Similar reasons for nonintervention were captured for women and men (50% due to comorbidities alone, 36% morphology and comorbidity). The endovascular repair imaging analysis revealed no gender differences. Among truly untreated women, ruptures were common (18%), and mortality was high (86%).

CONCLUSIONS

Surgical AAA management differed between women and men. Women could be underserved in terms of elective repairs: one in every four women was untreated with over-the-threshold AAAs. The lack of obvious gender differences in eligibility analyses could imply unmeasured discrepancies (eg, in disease extent or patient frailty).

Gender Bias in Judging Frailty and Fitness for Lung Surgery

BACKGROUND

Disparities in surgical care for lung cancer have been well documented, and unconscious bias may be a source of inequity. We assessed whether gender biases exist when nonclinical decision makers render decisions about major lung surgery.

METHODS

Amazon Mechanical Turk workers, remotely located "crowdworkers" readily available for hire to perform discrete on-demand tasks on the Amazon Mechanical Turk platform, were each shown 4 videos of different standardized patients (SPs) in a clinic setting, 1 video in each energy level (vigorous or frail) and race category (White or Black), randomized to male or female. Workers scored video characteristics and whether they would support the SP's decision to undergo a major lung operation.

RESULTS

A total of 855 workers were recruited. The frail White male SP was more likely to have support to undergo lung surgery than the frail White female SP, while the frail Black male SP was much less likely to have support to undergo lung surgery than the frail Black female SP. There were no significant differences in support for surgery between the vigorous male and female SPs and ratings by male and female workers in their recommendations.

CONCLUSIONS

Biases related to patient gender exist in the general population and affect views on surgery, particularly in the setting of frailty. Understanding such differences may aid in educational efforts directed at reducing gender-based biases in treatment recommendations.

CT-Derived Body Composition Values and Complications After Pneumonectomy in Lung Cancer Patients: Time for a Sex-Related Analysis?

Purpose

This study aimed to assess if CT-derived body composition values and clinical characteristics are associated with the risk of postsurgical complications in men and women who underwent pneumonectomy for lung cancer.

Materials and Methods

Patients who underwent pneumonectomy between 2004 and 2008 were selected. The ethics committee approved this retrospective study with waiver of informed content. Main clinical data collected were sex, age, weight and height to calculate body mass index (BMI), albumin, C-reactive protein, smoking status, side, sarcopenia, presurgical treatments, reoperation, and complications within 30 days after pneumonectomy, classified as: lung complications, cardiac complications, other complications, and any complication. From an axial CT image at the level of L3, automatic segmentations were performed to calculate skeletal muscle area (SMA), skeletal muscle density, subcutaneous adipose tissue, and visceral adipose tissue. Skeletal muscle index was calculated as SMA/square height. Univariate and multivariate logistic regression analyses were performed to estimate the risk of any complication, both on the total population and in a by sex subgroup analysis. All tests were two tailed and considered significant at 5% level.

Results

A total of 107 patients (84 men and 23 women) were included. Despite no significant differences in BMI, there were significant differences of body composition values in muscle and adipose tissue parameters between men and women, with women being significantly more sarcopenic than men (p = 0.002). Separate analyses for men and women showed that age and SMA were significantly associated with postoperative complications in men (p = 0.03 and 0.02, respectively).

Conclusions

Body composition measurements extracted from routine CT may help in predicting complications after pneumonectomy, with men and women being different in quantity and distribution of muscle and fat, and men significantly more prone to postpneumonectomy complications with the increase of age and the decrease of skeletal muscle area.

Definition and assessment of high risk in patients considered for lobectomy for stage I non-small cell lung cancer: The American Association for Thoracic Surgery expert panel consensus document

OBJECTIVE

Lobectomy is a standard treatment for stage I non-small cell lung cancer, but a significant proportion of patients are considered at high risk for complications, including mortality, after lobectomy and might not be candidates. Identifying who is at risk is important and in evolution. The objective of The American Association for Thoracic Surgery Clinical Practice Standards Committee expert panel was to review important considerations and factors in assessing who is at high risk among patients considered for lobectomy.

METHODS

The American Association for Thoracic Surgery Clinical Practice Standards Committee assembled an expert panel that developed an expert consensus document after systematic review of the literature. The expert panel generated a priori a list of important risk factors in the determination of high risk for lobectomy. A survey was administered, and the expert panel was asked to grade the relative importance of each risk factor. Recommendations were developed using discussion and a modified Delphi method.

RESULTS

The expert panel survey identified the most important factors in the determination of high risk, which included the need for supplemental oxygen because of severe underlying lung disease, low diffusion capacity, the presence of frailty, and the overall assessment of daily activity and functional status. The panel determined that factors, such as age (as a sole factor), were less important in risk assessment.

CONCLUSIONS

Defining who is at high risk for lobectomy for stage I non-small cell lung cancer is challenging, but remains critical. There was impressive strong consensus on identification of important factors and their hierarchical ranking of perceived risk. The panel identified several key factors that can be incorporated in risk assessment. The factors are evolving and as the population ages, factors such as neurocognitive function and frailty become more important. A minimally invasive approach becomes even more critical in this older population to mitigate risk. The determination of risk is a clinical decision and judgement, which should also take into consideration patient perspectives, values, preferences, and quality of life.

Association of patient gender and gastroenterologists' diagnosis and management choices in gastroesophageal reflux disease

Symptom severity and prevalence of erosive disease in gastroesophageal reflux disease (GERD) differ between genders. It is not known how gastroenterologists incorporate patient gender in their decision-making process. We aimed to evaluate how gender influences the diagnosis and management recommendations for patients with GERD. We invited a nationwide sample of gastroenterologists via voluntary listservs to complete an online survey of fictional patient scenarios presenting with different GERD symptoms and endoscopic findings. Patient gender for each case was randomly generated. Study participants were asked for their likelihood of a diagnosis of GERD and subsequent management recommendations. Results were analyzed using chi-square tests, Fisher Exact tests, and multivariable logistic regression. Of 819 survey invitations sent, 135 gastroenterologists responded with 95.6% completion rate. There was no significant association between patient gender and prediction for the likelihood of GERD for any of the five clinical scenarios when analyzed separately or when all survey responses were pooled. There was also no significant association between gender and decision to refer for fundoplication, escalate PPI therapy, or start of neuromodulation/behavioral therapy. Despite documented symptomatic and physiologic differences of GERD between the genders, patient gender did not affect respondents' estimates of GERD diagnosis or subsequent management. Further outcomes studies should validate whether response to GERD treatment strategies differ between women and men.

The role of gender in non-small cell lung cancer: a narrative review

The role of gender in the development, treatment and prognosis of thoracic malignancies has been underappreciated and understudied. While most research has been grounded in tobacco-related malignancies, the incidence of non-smoking related lung cancer is on the rise and disproportionately affecting women. Recent research studies have unveiled critical differences between men and women with regard to risk factors, timeliness of diagnosis, incongruent screening practices, molecular and genetic mechanisms, as well as response to treatment and survival. These studies also highlight the increasingly recognized need for targeted therapies that account for variations in the response and complications as a function of gender. Similarly, screening recommendations continue to evolve as the role of gender is starting to be ellucidated. As women have been underrepresented in clinical trials until recently, the data regarding optimal care and outcomes is still lagging behind. Understanding the underlying similarities and differences between men and women is paramount to providing adequate care and prognostication to patients of either gender. This review provides an overview of the critical role that gender plays in the care of patients with non-small cell lung cancer and other thoracic malignancies, with an emphasis on the need for increased awareness and further research to continue elucidating these disparities.

Economics influences therapy decisions in chronic myeloid leukaemia: should it?

The question whether economic considerations should influence therapy decisions in persons with chronic myeloid leukaemia (CML) is complex touching on many metrics other than medicine including the perceptions, attitudes, and motivations of physicians, patients, their families, and payors', allocation of health care resources, and others. We discuss metrics whereby physicians, patients, and payors make CML therapy choices in settings where cost and availability are or are not considerations. We conclude that economic considerations strongly influence therapy decisions in CML. Whether this should be so and what impact it has on outcomes is also considered. Absent definitive data proving which therapy strategy is best allowing economic considerations to operate may not be as unreasonable or unethical as it appears. In some settings, it may be the best approach. However, because TKIs markedly prolong survival and may even cure some persons with CML, TKI therapy should be available to everyone with CML.

Disparities in neoadjuvant radiation dosing for treatment of rectal cancer

BACKGROUND

Certain patients are less likely to undergo appropriate cancer treatment, worsening their overall cancer survival. The purpose of this investigation was to identify factors associated with inadequate neoadjuvant radiation for rectal cancer.

METHODS

The National Cancer Database was queried for patients with locally advanced rectal cancer who received neoadjuvant radiation 2006-2014. Adequate radiation was considered to be 4,500-5,040 cGy. Demographic, hospital and clinical variables were analyzed for association with inadequate radiation.

RESULTS

The study cohort was 34,391 patients; 1,842(5.4%) received inadequate radiation. On multivariate analysis, female gender, older age, other race, government-provided insurance, lower clinical stage and rural location correlated with inadequate radiation.

CONCLUSIONS

Women were 50% less likely than men to receive correct neoadjuvant radiation dosing. Other factors including age, race, insurance, clinical stage, geographic location and neoadjuvant chemotherapy were significantly associated with radiation dosing. These factors should be evaluated to determine if they can be modified to improve outcomes.

Is there any gender or age-related discrepancy in the waiting time for each step in the surgical management of acute traumatic cervical spinal cord injury?

Context/Objective: Prior studies indicate that patient's gender and age can influence treatment choices during spine disease management. This study examines whether individual's gender and age at injury onset influence the waiting time for each step in the surgical management of patients with acute traumatic cervical spinal cord injury (atcSCI). Design: Retrospective cohort study. Setting: Quaternary spine trauma center. Participants: This study included consecutive individuals with atcSCI admitted from August/2002 to October/2008 who were enrolled in the Surgical Trial in Acute Spinal Cord Injury Study (STASCIS). Interventions: Spinal cord decompression. Outcome Measures: Data on the periods of time for each step in the surgical management were analyzed to explore the potential effects of gender and age at injury onset. Results: There were 64 individuals with atcSCI (17 women, 47 men; age range: 18-78 years; mean age: 50.5 ± 2.1 years). Older age was associated with longer stay in the acute spine center, but this association was cofounded by major pre-existing medical co-morbidities. Age did not significantly affect the waiting time for each step in the surgical management of these individuals with atcSCI. Women underwent surgical assessment earlier than men. Gender did not influence other key steps in the surgical management. Conclusion: The study results suggest that older age at injury onset was associated with longer stay in the acute spine care center, and women had a shorter waiting time for surgical assessment than men. Nevertheless, no other age or gender bias was identified in the waiting times for the steps in the management of atcSCI.

Gender Bias Affects Assessment of Frailty and Recommendations for Surgery

BACKGROUND

Physician gender bias in surgical treatment recommendations is recognized but not well understood. This study hypothesized that gender differences may exist in interpretation of patients' physical behaviors and that these differences may be associated with decision making by providers and surrogate decision makers.

METHODS

A pool of Amazon Mechanical Turk workers was solicited to participate in an online assessment. Workers viewed 3 short videos of standardized patients (SPs) trained to exhibit physical characteristics of vigorous, frail, and neither vigorous nor frail (average) behavior and then answered survey questions related to video characteristics and whether they would support the SP's decision to undergo an indicated major lung resection.

RESULTS

There were 724 participating workers; their mean age was 42.6 ± 11.8 years, and 386 were women. Men judged the average SP to be younger (P = .025), and women were more likely to recognize weight loss in the frail SP (P = .009). Overall, men and women were equally supportive of lung resection when indicated. The likelihood of supporting a decision to proceed with resection was inversely related to SP distress (P < .001) and was directly related to increasing gait speed (P < .001), energy (P < .001), and strength (P < .001). Male participants were less likely to support resection related to higher energy (P = .02) and strength levels (P = .016).

CONCLUSIONS

Gender differences exist in how video portrayal of patient frailty is perceived and affects surgical recommendations. Understanding such differences may aid in educational efforts directed at reducing gender-based biases in treatment recommendations by physicians and surrogate decision makers.

Trends in Sentinel Lymph Node Biopsy Enactment for Cutaneous Melanoma

BACKGROUND

Over recent years, sentinel lymph node biopsy (SLNB) recommendations in guidelines for cutaneous melanoma have changed considerably. We aimed to assess trends in enactment of SLNB to evaluate to what extent guidelines were adhered to, and to identify clinical and pathological determinants of (non-)adherence.

METHODS

Clinicopathological data from the Dutch nationwide network and registry of histopathology and cytopathology were retrieved from patients diagnosed with primary cutaneous melanoma in The Netherlands between 2003 and 2014. SLNB enactment was analyzed per year. Multivariable regression models were developed to assess the determinants of SLNB enactment.

RESULTS

A total of 51,510 primary cutaneous melanomas in 49,514 patients were diagnosed, of which 24,603 melanomas were eligible for SLNB as they were staged T1b or higher. In practice, only 9761 (39.7%) patients underwent SLNB, with an increasing trend from 39.1% in 2003 to 47.8% in 2014 (p < 0.001). A total of 759 (2.9%) of 26,426 patients without SLNB indication underwent SLNB anyway. Variables significantly associated with enactment of SLNB were male sex, younger age, and melanoma on sites other than the head and neck.

CONCLUSIONS

Although there was an increasing trend in time in SLNB enactment, enactment of SLNB did not comply well with recommendations in (inter)national guidelines. Female sex, higher age, and melanoma located on the head and neck were associated with non-enactment of SLNB.

Physician Gender Differences in Processing Surgical Risk Features in Videos of Standardized Patients

BACKGROUND

Physician gender differences exist for estimating surgical risk and recommending lung resection. This study assessed gender differences in perceived importance of health characteristics portrayed by standardized patients posing as lung resection candidates.

METHODS

Physicians read a clinical vignette categorized as low, average, or high surgical risk and then viewed a video of a standardized patient exhibiting vigorous, normal, or frail behavior. The relative importance of gait speed, strength, fatigue, age, and weight loss in the videos was scored on a five-point Likert-like scale. Ratings of the importance of each were compared by gender and risk category.

RESULTS

Of 73 participating physicians, 62 were male and 11 were female, 40 were thoracic surgeons, and 33 were cardiothoracic surgical trainees. All video features were scored as very important or somewhat important a majority of the time. Gait speed and strength ratings were strongly correlated (r = 0.76), followed by strength and fatigue (r = 0.52) and gait speed and fatigue (r = 0.51). Female gender was associated with significantly greater odds of rating age as very important (p = 0.040). For weight loss, the differences between genders varied significantly (females consistently rated weight loss as important, whereas ratings for men varied by risk category; p = 0.002 for the interaction).

CONCLUSIONS

Physicians variably rate certain health characteristics of standardized patients as important in making surgical risk estimations. Women and men rate the importance of age and weight loss differently. These findings may help educate physicians to develop more consistent estimates of surgical risk.

Do Estimates of Treatment Risk Based on Clinical Vignettes Differ by Physician Gender?

BACKGROUND

Clinical vignettes are frequently used as instructional and evaluative instruments for physicians. Physicians' gender is a source of unconscious bias in treatment recommendations. This study assessed whether interpretation of information in clinical vignettes differed by physicians' gender as a possible source of unconscious bias.

METHODS

Thoracic surgeons and physicians in cardiothoracic surgical training were asked to provide estimates of major complication rates for lung resection on the basis of anonymized clinical vignettes of patients undergoing lung resection. Vignettes were categorized as low, average, and high risk by using a sum of Charlson Comorbidity Index (possible range, 0 to 37) and a combined physiologic score, EVAD (forced expiratory volume in 1 second, diffusing capacity of lung for carbon monoxide, age; possible range, 0 to 12); participants were not aware of the risk scores or vignette categories. Generalized estimating equation linear regression models were fit with risk scores treated as a continuous independent variable.

RESULTS

A total of 247 physicians (105 practicing surgeons, 142 trainees; 203 men, 44 women) participated in one or more of the studies. Nearly all (103; 98%) of the practicing surgeons rated themselves as competent or expert in lung resection compared with 77 (54%) of the trainees (p < 0.001). Participants' complication estimates mirrored both vignette risk category and combined risk score. There was no significant difference between men and women physicians in their estimates of complication rates.

CONCLUSIONS

Unconscious bias related to physicians' gender is not associated with differential use of information in clinical vignettes. Any possible bias may arise from face-to-face interactions with patients. Research into physicians' and patients' gender differences during such interactions is warranted.

Does Race Influence Risk Assessment and Recommendations for Lung Resection? A Randomized Trial

BACKGROUND

Racial disparities in use of surgical therapy for lung cancer exist in the United States. Videos of standardized patients (SPs) can help identify factors that influence physicians' surgical risk estimation. We hypothesized that physician race and SP race in videos influence surgeon decision making.

METHODS

Four race-neutral clinical vignettes representing lung resection candidates were paired with risk-level concordant short silent videos of SPs. Vignette/video combinations were classified as low or high risk. Trainees and practicing thoracic surgeons read a race-neutral vignette, provided an initial estimate of the percentage risk of major surgical complications, viewed a video randomized to a black or white SP, provided a final estimate of risk, and scored the likelihood that they would recommend operative therapy. Changes in risk estimates were assessed.

RESULTS

Participants included 113 surgeons (38 practicing surgeons, 75 trainees); of these, 76 were white non-Hispanic (67%), and 37 were other self-identified racial categories. Percentage changes between initial and final risk estimates were not significantly related to patient race (p = 0.11) or surgeon race (white versus other; p = 0.52). Videos of black SPs were associated with a similar likelihood of recommending an operation compared with that of videos of white SPs (p = 0.90). Physician race (white versus other) was not related to the likelihood of recommending surgical intervention (p = 0.79).

CONCLUSIONS

Neither patient nor physician race was significantly associated with risk estimation or surgical recommendations. These findings do not provide an explanation for documented racial disparities in lung cancer therapy. Further investigation is needed to identify the mechanism underlying these disparities.