Laparoscopy has revolutionized the field of surgery and is increasingly becoming the standard of care for many routine intra-abdominal operations. Inguinal hernia repair, cholecystectomy, appendectomy, and colectomy are among the most common surgeries performed in a routine general surgical practice and are increasingly performed using minimally-invasive techniques. Recent meta-analyses and retrospective studies have shown that patients who receive a minimally-invasive approach to these four common surgeries have decreased post-operative pain, earlier return to normal activities, and reduced incidence of post-operative complications such as wound infection, when compared to patients who undergo similar open surgery [1,2,3,4,5,6,7]. Relative to an open approach, a laparoscopic approach requires additional surgical skill, training, and more expensive equipment. Despite the fact that an open approach to operations of the abdomen is typically performed by most surgeons and reduces operation length, laparoscopic procedures require additional skills and are less invasive, improving the patient’s recovery experience.

Laparoscopic surgeries gain access to the abdomen via tiny, strategically placed ports to avoid important vessels and nerves [8, 9]. While this leads to reduced visibility and tactile feedback for the surgeon, the smaller incisions result in lower incidence of surgical site infection, bleeding, scarring, and post-operative pain. In turn, laparoscopic surgeries are met with improved patient satisfaction [10]. As such, an open approach to surgery is now reserved only for patients presenting more advanced or complicated cases (i.e., an unusually large hernia, severe cholecystitis) or a potentially compromising pre-existing condition (i.e., cancer, abnormal anatomy) or to convert intra-operatively to avoid or treat a complication from laparoscopic approach. Otherwise, the standard of care has shifted toward a laparoscopic procedure for many intra-abdominal operations [1,2,3,4,5,6,7].

Current literature has identified racial disparities in the quality of care received by patients in a variety of healthcare settings. While there is a lack of literature describing racial disparities in access to minimally-invasive surgery, racial disparities in clinical care is well-documented [11, 12]. Risk factors for decreased quality of care that are currently under investigation include patient race, ethnicity, gender, and weight. Previous investigation into disparities in access to basic laparoscopic surgery has been limited in that only non-profit academic medical centers were included for analysis [13]. The present study examines the influence of race on access to minimally-invasive surgery for patients in need of inguinal hernia repair, cholecystectomy, appendectomy, and colectomy utilizing a database with participation from governmental, municipal, academic, and community hospitals. We hypothesized that race may be significantly associated with access minimally-invasive general surgical procedures.

Materials and methods

Data collection

The American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) and the hospitals participating in the ACS-NSQIP are the source of the data used herein; they have not verified and are not responsible for the statistical validity of the data analysis or the conclusions derived by the authors.

This was an IRB-approved retrospective study utilizing data from the ACS-NSQIP database. ACS-NSQIP is a source of more than 150 prospectively recorded demographic, pre-operative, intra-operative, and 30-day post-operative variables. Data are collected on randomly-selected patients in a HIPAA-compliant manner by dedicated clinicians. In 2016, 680 hospitals participated in ACS-NSQIP, submitting data for more than 1 million cases. The database was queried to identify cases of individuals who underwent inguinal hernia repair, cholecystectomy, appendectomy, or colectomy in 2016. The following CPT codes corresponded to open approach to surgery: 44950, 47,612, 44,140, and 49,505. The following CPT codes corresponded to laparoscopic approach to surgery: 44,970, 47,562, 44,204, and 49,650.

Variables of interest

Variables recorded from ACS-NSQIP included demographic and pre-operative clinical covariates. Demographic variables included race and ethnicity (white, black, American Indian/Alaska Native, Asian Pacific Islander, Hispanic, and unknown), age, and sex. Pre-operative clinical variables that the attending surgeon would have access to were selected for analysis, including American Association of Anesthesiologists (ASA) class, current smoking status, chronic obstructive pulmonary disease (COPD), body mass index (BMI), diabetes, congestive heart failure (CHF), medicated hypertension, renal failure or dialysis, steroid use, and bleeding disorders. All subsequent statistical analyses were performed using SAS 9.4 (Cary, NC).

Univariate analyses

The primary outcome of interest was laparoscopic approach to surgery. Descriptive statistics were performed to characterize the patients included in this study (Table 1). Univariate analyses were then performed to identify demographic and pre-operative variables associated with the primary outcome. Chi squared tests and Student’s t-tests were used for categorical and continuous variables, respectively, with p values < 0.05 being regarded as statistically significant. Chi squared tests were done on all procedures to elucidate the association between all races and ethnicities listed in ACS-NSQIP and laparoscopic approach to surgery (Table 2).

Table 1 Demographic and clinical characteristics of whole patient cohort from ACS-NSQIP
Full size table
Table 2 Demographic and clinical characteristics of propensity score matched groups
Full size table

Propensity score matching

To eliminate undesired bias associated with racial differences in comorbidities and identify demographic and pre-operative variables independently associated with laparoscopic approach to the surgical procedures under investigation, a propensity score matching algorithm was utilized. The PSMATCH procedure was used to develop a 1:1 greedy-matching algorithm [14] that included variables that the attending surgeon would have access to pre-operatively and inguinal hernia repairs were matched for unilateral and bilateral (Table 1). Chi squared tests were then implemented to confirm group balancing and determine the association between race and access to minimally-invasive approach to surgery (Table 2). p values < 0.05 were regarded as statistically significant.

Results

Patient characteristics

There were 41,340, 3182, 60,444, 50,523, and 58,012 cases of unilateral inguinal hernia repair, bilateral inguinal hernia repair, cholecystectomy, appendectomy, and colectomy identified, respectively. Of these patients, 67.53, 70.33, 60.31, 59.56, and 70.32% self-reported as white; 8.59, 5.59, 8.76, 6.69, and 9.02% self-reported as black; 0.44, 0.57, 0.68, 0.58, and 0.43% self-reported as American Indian/Alaska native; 2.18, 1.79, 3.39, 4.50, and 2.63% self-reported as Asian/Pacific Islander; 7.38, 7.13, 13.51, 13.00, and 5.41% self-reported as Hispanic; and 13.88, 14.58, 13.34, 15.66, and 12.19% patients were of unknown race/ethnicity, respectively (Table 1).

Overall, patients of non-white race/ethnicity had a lower incidence of minimally-invasive approach to surgery. With respect to unilateral inguinal hernia repair, bilateral inguinal hernia repair, cholecystectomy, appendectomy, and colectomy, 26.61, 62.02, 94.90, 94.88, and 54.47% of white patients; 20.10, 53.37, 94.28, 92.81, and 48.60% of black patients; 16.48, 66.67, 95.38, 95.88, and 55.02% of American Indian/Alaska native patients; 29.41, 59.65, 91.52, 91.25, and 59.19% of Asian/Pacific Islander patients; 20.33, 56.83, 96.40, 95.13, and 53.36% of Hispanic patients; and 14.54, 43.10, 95.44, 91.65, and 50.97% of patients of unknown race/ethnicity underwent a laparoscopic approach to their surgery (Table 1). On Chi squared tests conducted on the entire patient cohort and across all four surgeries, there was a significant association between self-reported race/ethnicity and access to laparoscopic approach to surgery (laparoscopic approach on 95.88% of American Indian/Alaska Native patients, 91.25% of Asian/Pacific Islander patients, 92.81% of black patients, 95.13% of Hispanic patients, 94.88% of white patients, and 91.65% of patients of unknown race/ethnicity, p < 0.0001).

Propensity score matching

After propensity matching, there were 7102, 354, 10,574, 6758, and 10,470 cases of unilateral inguinal hernia repair, bilateral hernia repair, cholecystectomy, appendectomy, and colectomy, respectively. Groups were balanced for all covariates with the exception of race; half of each patient cohort self-reported as white while the other half self-reported as black. On univariate analysis, black race was significantly associated with undergoing an open approach in all four surgical procedures under investigation, (Table 2) (41.13% of white people and 58.87% of black people, p < 0.0001 for unilateral inguinal hernia repair; 35.16% of white people and 64.84% of black people, p < 0.0001 for bilateral inguinal hernia repair; 97.98% of white people and 94.29%, p < 0.0001 of black people for cholecystectomy; 98.85% of white people and 92.81% of black people, p < 0.0001 for appendectomy; and 70.93% of white people and 48.60% of black people, p < 0.0001 for colectomy).

Discussion

The impact of racial disparities in access to minimally-invasive surgery

In this study, we found that black, Hispanic, and Asian patient populations requiring inguinal hernia repair, cholecystectomy, appendectomy, and colectomy were less likely to receive minimally-invasive surgery via laparoscopy compared to the white patient population. As previously mentioned, an open approach to these procedures is justified and safe for patients presenting with a complex case or a potentially compromising pre-existing condition or to intra-operatively save a patient from a laparoscopic complication. However, an open approach to surgery is avoided when possible because of higher rates of surgical site infection, longer hospital stays, increased bleeding and hemorrhaging, and more scarring [10]. But ultimately, the decision is made by the attending surgeon.

Decreased access to minimally-invasive general surgery for minority groups is one of many ways in which racial disparities in healthcare affect patient care. Racial disparity in healthcare quality is a complex, multi-faceted, and national problem. And of greatest relevance to the present study, it has been reported that racial disparities in general surgical outcomes exist, although the etiologies of these disparities have not been elucidated [15, 16]. In the pursuit of healthcare equity, elucidating the social, economic, and clinical factors that generate racial disparities is essential. There are several theories that may explain racial disparities in access to laparoscopic approach to inguinal hernia repair, cholecystectomy, appendectomy, and colectomy. Some of the theories suggested by researchers are: economic barriers in access to standard of care, geographic differences in standard of care, and the impact of implicit bias in clinical practice.

Economic barriers in access to standard of care

In the racial disparity identified herein, it may be that race is a proxy for socioeconomic and insurance status. The propensity score matching algorithm used in the present study did not balance for insurance coverage because the intent of the authors was to control only for clinical variables directly related to the presentation of a patient’s condition. Therefore, the racial disparity in access to minimally-invasive inguinal hernia repair, cholecystectomy, appendectomy, and colectomy identified in this study may be in part due to economic barriers related to insurance coverage.

Currently, reimbursement rates are determined via a complex system that takes resources, physician work, time commitment, and malpractice expense into account [17]. There exist significant differences in this system among the public and private health insurance infrastructures. Consequently, hospitals and physicians are reimbursed at different rates depending on patient insurance coverage. In this way, reimbursement policy may potentially play a role in clinical decision making, forcing a surgeon to make a clinical decision that is in lieu with an economic situation or hospital practice. In particular, academic surgery departments are heavily influenced by such factors, as their ability to fund residency and medical student education is strongly driven by surgical reimbursements [18]. Furthering complicating this issue, a study by Hoballah et al. [18] reported that reimbursement rates for general surgical procedures steadily decreased from 1960 to 2006 despite the concurrent growth the healthcare industry.

Minimally-invasive surgery requires a higher level of technical skill, more time in the operating room, and costlier equipment and supplies. Despite this, current public insurance payment structures do not necessarily reflect this and often reimburse surgeons less for performing minimally-invasive surgery. Surgeons are therefore sometimes faced with a financial disincentive in choosing a laparoscopic approach to an operation [19, 20]. Although reimbursement rates would ideally not affect clinical decision making, current literature demonstrates that this is not the case. Low reimbursement rates not only may cause an inclination toward or preference for performing surgeries by an open approach but may also financially disincentive surgeons with little background in laparoscopy to work at becoming proficient in methods of minimally-invasive surgery. Angus et al. [21] has shown that physicians are faced with a financial disincentive in performing laparoscopic Roux-en-Y gastric bypass on patients with morbid obesity covered by public insurance. This pushed bariatric surgeons to perform open gastric bypass surgery on a patient that may otherwise be an appropriate candidate for laparoscopic Roux-en-Y gastric bypass surgery. Public healthcare reimbursement rates may also disincentivize certain hospitals to promote minimally-invasive surgery. Frazee et al. [22] found that low-volume hospitals struggle to achieve positive financial margins for laparoscopic cholecystectomy under Medicaid reimbursement rates, which may negatively impact the quality of care received by populations who use such hospitals.

The consequences of coverage by insurance that reimburses physicians at relatively low rates may disproportionately affect minority patients. According to Medicaid Enrollment by race and ethnicity as of 2013, Medicaid recipients are 40% white, 21% black, 25% Hispanic, and 14% as others. Given the relative disincentive for surgeons to treated publicly-insured patients, it may be that disparities in quality of care received by patients of minority backgrounds begin in the economic differences generated by medical insurance. Research has shown that uninsured and publicly-insured individuals have higher morbidity and mortality rates [23,24,25,26,27]. Because minority patients are more likely to fall into these insurance categories, they may face higher morbidity and mortality rates. But importantly, this trend has also been shown to persist when insurance status is controlled for [28].

Geographic differences in access to standard of care

Utilization of minimally-invasive surgery varies in different regions of the United States. Cooper et al. [29] investigated this geographic variance by querying the United States nationwide inpatient sample database to identify all patients who underwent appendectomy, colectomy, total abdominal hysterectomy, and lung lobectomy in 2010. In the case of colectomy and appendectomy, this study found that minimally-invasive surgery is more likely to be conducted in urban locations, teaching hospitals, large hospitals, and the Midwest. This finding underscores the import of the hospital itself in influencing patient access to minimally-invasive surgery. Moreover, Cooper et al. [29] emphasized that certain hospitals have a more minimally-invasive culture, whereas others have a more open surgery culture.

Individuals of minority race and/or ethnicity tend to live in urban areas where there are a variety of teaching and community hospitals to choose from. But this may not necessarily translate into flexibility in hospital choice, as black individuals may likely to be treated at low-volume hospitals [30, 31], which have been shown to struggle to provide similar access to laparoscopic surgery due to current reimbursement models [22]. Indeed, several studies on racial disparities in access to high-quality surgical care have found that the phenomenon may be due to hospital-level factors. In cardiac [32], vascular [33], emergency [34], and trauma surgery [35], the risk of mortality among non-white patients has been shown to be positively correlated with the proportion of non-white patients treated a hospital, suggesting that the institutions predominantly responsible for providing care to non-white populations tend to underperform. In more rural areas, which tend to serve populations with a smaller proportion of non-white individuals, it is likely that individuals needing surgery are all sent to one large, teaching hospital that serves as a regional medical center. As such, it may be that the racial disparities identified in this study were influenced by geography.

Implicit bias in healthcare

Lastly, one of the theories suggested by experts regarding the racial disparities in healthcare quality is the role of implicit bias and its influence on clinical decision making [11, 12, 36]. Implicit bias refers to an unintentional and unrecognized preference for one group over another. Because implicit biases necessarily exist unacknowledged, an individual may explicitly believe in social equity but simultaneously have an inadvertent implicit bias against a given group. This implicit bias can affect social interactions with individuals who are members of the group against which the implicit bias is held [37]. It has been demonstrated that clinicians’ implicit biases and minority patients’ physician mistrust, which is largely resultant of institutional racism, are associated with poor patient-clinician communication [38,39,40,41,42]; which may translate into inferior patient outcomes and quality of care. Researchers have found that this may in fact be the case, as it has been shown that physicians’ implicit biases are associated with outcome differences [39,40,41, 43].

A study by Borkhoff et al. [44] demonstrated that implicit biases can be held against demographic groups other than race. In this study, a group of orthopedic surgeons and family practitioners were provided with radiograph images revealing osteoarthritis coupled with patient vignettes describing moderate unilateral knee pain. When the vignettes were ascribed to a male patient, orthopedic surgeons were 22 times more likely to recommend total knee arthroplasty and family practitioners were 2 times more likely to recommend total knee arthroplasty. Furthermore, Schwartz et al. [45] investigated the prevalence of pro-thin bias in physicians. When given images of obese and non-obese people, healthcare providers were more likely to associate obese people with negative qualities (provided the following three stereotypes: lazy-motivated, smart-stupid, and valuable-worthless) even if their work emphasized obesity research or clinical care. Our investigation utilized a propensity score matching algorithm that balanced for patient gender and BMI. Given the compelling existing research suggesting the prevalence of pro-male and pro-thin bias among physicians, the authors recognize the need for further investigation into the influence of these factors to independently influence access to minimally-invasive surgery.

Heightened insight into the depth of demographic disparities in clinical care has elucidated a widespread need to address the issue of implicit bias in the medical community [12, 46]. It has been shown that physicians of all races demonstrate measurable pro-white bias, as determined by the Race Attitude Implicit Association Test (IAT), a widely-used measure of hidden bias that has been shown to reliably predict behavior [47, 48]. Several studies have shown the impact of implicit bias on clinical decision making. This phenomenon is particularly well studied in cardiovascular disease. Compared to white patients, black patients are less likely to be diagnosed with thrombolysis or indicated for coronary artery bypass surgery and less invasive cardiovascular surgeries [49,50,51].

Understanding the source of disparities in clinical decision making is complex. Some researchers have speculated that the thorough scientific training received by physicians may strengthen their confidence in their own objectivity, which can in turn cause these individuals to be more likely to act upon biases [52]. It has also been suggested that the time pressure faced by physicians in the clinical setting increases their susceptibility to unconscious bias [12, 53]. Similarly, clinical uncertainty may push doctors to inadvertently rely of inferences about the patient rather than evidence about the patient [36]. The findings herein underscore the far-reaching influence of implicit bias in healthcare, in that they clearly show outcome disparities on the basis of race alone.

Racial disparities in access to other high cost minimally-invasive surgeries

In an effort to contextualize the findings herein, a literature search was conducted to identify current understanding of racial disparities in access to a laparoscopic approach to surgeries that, like those investigated in this study, are high in volume and cost. Bariatric surgery, a widely accepted and scientifically proven method for control of both morbid obesity and obesity-related medical conditions, is one type of surgery where significant disparities according to race, income, education level, and insurance type continue to exist [54]. Similar disparities have been seen in endovascular procedures for peripheral arterial disease. Loja et al. [55] conducted an evaluation of California hospital systems between 2005 and 2009 and noted racial disparities in outcomes of treatment of peripheral arterial disease. The study concluded that Hispanic and black patients had worse amputation-free survival than non-Hispanic whites following endovascular therapy. Additionally, Hispanic and black patients were more likely to undergo lower extremity arterial reinterventions than non-Hispanic white patients. These are just two specific examples that demonstrate difference in access to minimally-invasive procedures in the surgical fields.

The authors have proposed that economic, geographic, and provider-level mechanisms contribute to the racial disparity in access to laparoscopic surgery. We propose that similar factors contribute to racial disparities across general surgery; however, further investigation is required.

Future outlook: minimizing demographic disparities in healthcare

There exists a clear need to address the root of evident racial disparities in patient diagnoses, treatments, and outcomes and promote clinical decision making that is truly evidence-based and not informed by heuristics. It has been proposed by investigators that the United States ought to implement what is known as a value-based payment system, which provide higher reimbursement rates for procedures that are associated with better patient outcomes [20]. A value-based payment system would include patient reported outcome measures in determining which procedures to incentivize via a higher reimbursement rates. While patient reported outcome measures are qualitative in nature, such measures can be effectively and accurately used and their incorporation into payment reform is widely supported by members of the healthcare industry who conduct research on the relationship between payment and quality of care [56, 57]. Further, a value-based payment system would also satisfy proponents of reimbursement rates founded in resource intensity, as minimally-invasive surgery requires more time and surgical skill [19]. Movement toward minimally-invasive surgery also presents financial benefit to insurance companies and health employers, as demonstrated by Keller et al. in the case of colectomy [58]. Despite the clear import of working to eliminate disparities in healthcare, revising current policy would doubtlessly be a lengthy and arduous task.

Addressing implicit bias in healthcare is equally complex. Perhaps the most logical time to work toward minimizing physician implicit bias is during medical training. Students matriculate with pre-established implicit biases, which have been shown to remain relatively unchanging throughout medical training [12]. Research suggests that simply confronting medical students with data demonstrating their implicit biases (i.e., a high IAT score) accomplishes little, because such data are at odds with self-perceived egalitarianism and is therefore regarded as a personal attack [59, 60]. It has been shown that underscoring the egalitarian goals of medical students prior to engagement in self-reflective discourse increases the likelihood that a student is open to discussing potential need to change his or her own biases [59]. However, awareness of one’s own implicit biases alone is not sufficient in truly altering the way an individual perceives others [12]. As such, intentional intervention is required.

Lai et al. [61] conducted comparative testing of 17 interventional methods for reducing implicit biases using IAT scores as a measure of intervention success. This study showed that exposure to counter stereotypical exemplars (i.e., a vignette in which a white man is violent and cruel and a black man is heroic), evaluative conditioning (in which stimuli are paired, such as “black” and “good”), and clearly outlined strategies for overcoming implicit biases are most effective. Devine et al. [62] developed a multi-faceted strategy that effectively reduced anti-black implicit bias among participants over 12-weeks. Here, all participants completed an IAT, ensuring awareness of their own implicit biases. This was followed by training sessions that provided participants with a wide array of bias-reducing strategies, intended to give participants flexibility. Participants also demonstrated increased concern for the undesirable consequences of racial bias [62]. Generally, researchers have advocated for intervention that is multi-faceted, longitudinal, and well-incorporated into medical education curriculum [61,62,63].

Further exploration is needed to develop an intervention that is both effective and durable [61, 64]. Additionally, some researchers have proposed that increasing the number of black physicians could reduce the incidence of anti-black bias in healthcare [12, 39], a claim that has been well-supported by investigation [47, 65, 66]. For example, Sabin et al. [47] found significant implicit bias amongst physicians of all racial groups with the exception of black physicians, who were neutral, and female physicians, who were found to demonstrate lower implicit bias than their male counterparts. Regardless of the manner in which implicit bias in healthcare are addressed, the path toward minimizing the effects of physician biases on clinical decision making will be complex and challenging.

Study strengths and limitations

Opportunities for training is an important topic that might explain some differences in access to minimally-invasive surgeries. Cooper et al. [29] suggested that variability in appropriate residency and fellowship training may be one reason that a given hospital underperforms minimally-invasive surgery. In a Canadian survey of a colorectal surgeons conducted in 2009, lack of adequate operative time and formal training were cited by surgeons as the main reasons for not performing laparoscopic colon resection [67]. Additionally, it is important to note that fellowship training in minimally-invasive surgery was not formalized until 1997 when minimally-invasive surgery fellowships began to develop. These fellowships were designed to train surgeons in advanced laparoscopic procedures and incorporate novel techniques such as single incision laparoscopic surgery into the skill set of attending surgeons with traditional training. It has been shown that the benefits of minimally-invasive surgery fellowship training improve outcomes for both complicated procedures such as laparoscopic Roux-en-Y gastric bypass and basic laparoscopic procedures such as appendectomies [68, 69]. While laparoscopic equipment may be readily available across all programs, there is lack of literature available demonstrating surgeon comfort level with basic laparoscopic procedures following residency training.

Of note, ACS-NSQIP does not provide specific hospital or physician identification. As a result, the potentially confounding effects of this variable could not be balanced by the propensity score matching algorithm used in the present study. Further, the propensity score matching did not balance for hospital location, and so the effect of geographic and provider-level characteristics on access to laparoscopic surgery is suggested because it is well-supported by current literature.

In the original data obtained from ACS-NSQIP, there were approximately ten-times more white patients than black patients in all four general surgical procedures. As a result, this study is limited in that many white patients had to be eliminated from analyses in order to implement propensity matching. However, the methodology utilized herein provides a statistically valid model to demonstrate the independent association between race and the surgical approach to abdominal operations. Lastly, because the data used in this study were obtained from ACS-NSQIP, the findings herein are not limited to a single institution and are therefore relevant to general surgical practices across the country.

Concluding remarks

There appears to be racial disparity in the application of a minimally-invasive approach to patients undergoing inguinal hernia repair, cholecystectomy, appendectomy, and colectomy. The authors propose that this disparity is a result of complex social, economic, and geographic factors. The findings herein warrant further investigation into the barriers preventing access to laparoscopic general surgical procedures that racial minorities might face. Additional work to determine if there exists a racial disparity in outcomes for patients undergoing these four surgical procedures needs to be performed.