Racial Differences in Planned Hysterectomy Procedure Route

Abstract

Background:

Hysterectomies can be performed with a minimally invasive surgical (MIS) approach or a laparotomic (abdominal) approach. The objective of this study was to assess any racial differences in the likelihood of having a planned MIS hysterectomy.

Materials and Methods:

A prospective cohort study of women undergoing hysterectomy at Henry Ford Health System was conducted where laparotomic and MIS approaches are available to all patients. All procedures were performed between October, 2015, and August, 2017. For this study, women were asked to report demographic and insurance information and complete validated questionnaires from 2 weeks before hysterectomy and up to six additional times in the year after hysterectomy. Clinical and operative characteristics were collected from electronic health records. Logistic regression and multinomial logistic regression models were applied to assess the association between race and the surgical approach.

Results:

Analyses included 235 White women and 196 Black women. Black women were less likely to have any MIS planned for their hysterectomy (odds ratio [OR] = 0.46, 95% confidence interval [CI] 0.3–0.71, p < 0.05), a laparoscopic hysterectomy (relative risk ratio [RRR] = 0.46, 95% CI 0.29–0.73, p < 0.05), or a vaginal hysterectomy (RRR = 0.45, 95% CI 0.25–0.81, p = 0.01) compared with White women. After adjusting for confounders, uterine weight and indication for surgery was fibroids, these racial differences did not remain statistically significant (MIS vs. abdominal [adjusted odds ratio {aOR} = 0.93, 95% CI 0.55–1.57, p = 0.79], laparoscopic vs. abdominal [adjusted relative risk ratio {aRRR} = 0.89, 95% CI 0.52–1.51, p = 0.54], and vaginal vs. abdominal [aRRR = 1.22, 95% CI 0.61–2.45, p = 0.58]). The associations were not confounded by the baseline survey data from standardized questionnaires on depression, financial distress, and satisfaction with their decision.

Conclusions:

Black women were not less likely than White women to have planned an MIS hysterectomy after controlling for important confounding variables. These results emphasize the importance of considering all important confounders when examining racial differences.

Introduction

Hysterectomy is the most commonly performed non-obstetric surgical procedure for women in the United States.¹ Hysterectomies can be performed with a minimally invasive surgical (MIS) approach (vaginally or laparoscopically with or without a robotic assist) or a traditional laparotomic approach (“open” abdominal). Recent analyses have suggested that risk of complications from surgery may be lower when hysterectomy is performed using an MIS approach rather than a laparotomic approach.^2,3 Prior work has also suggested that Black women may be less likely to have an MIS approach for their hysterectomy compared with White women. However, the results have varied, which could be due to differing study populations and variable inclusion of important confounders.^3

–8

The American College of Obstetricians and Gynecologists (ACOG) recommends that “Reducing racial and ethnic disparities in health and health care should be a priority for all obstetrician–gynecologists and other women's health care providers.”⁹ The ACOG guidelines suggest that this can be done by “raising awareness among colleagues, residents, staff, and hospital administrators about the prevalence of racial and ethnic disparities and the effect on health outcomes.” The goal of this work, based at a single health system in which open and MIS approaches are available to the entire patient population, was to assess whether there were any racial differences in the likelihood of having a planned MIS hysterectomy even after considering key potential confounders.

Materials and Methods

Data sources and study design

Henry Ford Health System (HFHS) is a vertically integrated health system with employed providers comprising its medical group, as well as a group of affiliated physicians who serve patients at HFHS hospitals. A subset of these physicians are trained and experienced in the performance of MIS hysterectomy. Within the HFHS medical group physician model, patients can be referred to another provider at HFHS who can perform an MIS hysterectomy for the patient. The referral process does not require the woman to establish herself as a new patient at a new clinic, which usually requires the patient to repeat reporting their health and surgical history, sign more the Health Insurance Portability and Accountability Act of 1996 (HIPAA) and insurance documentation, and ensure medical record transfer. If a HFHS patient is referred to another HFHS doctor for MIS, the woman will return to her original provider after the surgical course of care is completed.

The Detroit Hysterectomy Study is a prospective cohort study of women who had a hysterectomy between October, 2015, and August, 2017 at Henry Ford Hospital, Henry Ford West Bloomfield Hospital, or Henry Ford Macomb Hospital,^10,11 all of which offer MIS hysterectomy. The primary goal of this cohort study was to examine the effects of clinical characteristics on the outcomes of hysterectomy patients. The present analyses are secondary use of the data. This work was approved by the HFHS Institutional Review Board. Women scheduled for hysterectomy at the Henry Ford Hospitals who were at least 18 years of age and not having surgery for suspected cancer or with a concomitant urogynecology procedure were invited to participate in a longitudinal cohort study.¹⁰ The eligible women were called in the 2 weeks before their scheduled hysterectomy, defined as baseline (presurgery) interviews, and verbal informed consent was obtained.

For this study, women were asked to report demographic information (race, date of birth), insurance type (private/military or public [Medicare, Medicaid, or dual eligible for Medicare–Medicaid]), and complete validated questionnaires from baseline and up to six additional times in the year postsurgery, including 2 weeks before hysterectomy, 1 week, 4 weeks, 6 weeks, 3 months, 6 months, and 12 months after hysterectomy, on the following: pain scores (baseline current and average pain levels and postsurgery current pain levels due to their hysterectomy site or incision and apart from their incision [0 no pain–10 extreme pain]), the Patient Health Questionnaire-9 (PHQ-9) to assess depression (higher scores are greater severity of depression),¹² the Comprehensive Score for Financial Toxicity (lower scores suggest worse financial toxicity) to assess level of financial distress associated with having the hysterectomy,¹³ and the Decision Conflict Scale to assess personal uncertainty about having a hysterectomy.¹⁴ Satisfaction was defined as being “Satisfied” with their decisions using the Decision Conflict Scale where the women responded “Strongly Agree” or “Agree” with the statement—“I was satisfied with my decision.” We did not collect details on the surgical approaches discussed with the women or how the surgical plan was developed.

After surgery, participants' medical records were reviewed by Registered Healthcare Information Technicians for clinical and pre- and intraoperative details, including body mass index (BMI in kg/m²), uterine weight (in grams), all indications for surgery (uterine fibroids, endometriosis, pelvic pain, heavy bleeding, or other reasons), estimated blood loss (EBL in milliliters), intra- and postoperative transfusion, procedure duration (in minutes), length of hospital stay (in days), and whether the woman had any complications. Race was self-reported at the baseline presurgery interview. Only women who self-identified as Black/African American or White were included in these analyses as there were too few women in other race and ethnic groups for valid comparison. Age at date of verbal consent was calculated. The women who completed the baseline questionnaires and had hysterectomy details in their medical records and who self-reported their race as either Black or White were included in these analyses.

In this study, the primary outcome measure is the planned hysterectomy procedure route, which was defined by three mutually exclusive groups: (1) laparotomic/abdominal (“open”), (2) abdominal laparoscopic (with or without robotic assist), and (3) vaginal (with or without laparoscopy). All surgeries other than “open” abdominal were classified as MIS. If the planned MIS surgery was converted to an open procedure, the women were still classified as having their planned approach as MIS.

Statistical analyses

Pearson chi-square tests (categorical) and two-tailed t-tests (continuous) or one-way analysis of variance (continuous) were used to compare demographic, baseline presurgery, clinical, and operative characteristics between White women and Black women and between all surgical approaches. Post hoc tests were then conducted using chi-square pairwise comparisons for the categorical variables and pairwise t-tests with the Bonferroni method for the continuous variables. Odds ratios (ORs) and adjusted odds ratios (aORs) with 95% confidence intervals (CIs) were calculated from logistic regression models for the outcome of MIS versus non-MIS. We also used multinomial logistic regression models from which relative risk ratios (RRRs) and adjusted relative risk ratios (aRRRs) with 95% CIs were calculated to compare the racial difference among the three procedure routes (abdominal, abdominal laparoscopic, and vaginal).

Based on prior literature, the following variables in our data set were considered and assessed as possible confounders of the association between race (Black or White) and having a planned MIS: age, BMI, insurance type, baseline current pain level, baseline PHQ-9, baseline satisfaction, baseline financial toxicity, uterine weight (from the pathology report), and indication for surgery (fibroids, endometriosis, heavy bleeding, or pelvic pain). We then identified confounding factors from these selected variables if their addition to the regression model changed the OR for Black women versus White women of having the outcome by at least 10%.¹⁵

Missing values of possible confounders based on prior literature were imputed before building the regression models for identifying the confounders and testing the association between race and planned procedure route. During the imputation process using the multiple imputation chained equations technique, we included the outcome variable (planned hysterectomy procedure route), race, and the possible confounders mentioned before (age, BMI, insurance type, baseline current pain level, baseline PHQ-9, baseline satisfaction, baseline financial toxicity, uterine weight, and surgical indication) in the multiple imputation chained equations to impute the missing values of these possible confounders. Ten imputed data sets were generated with the predictive mean matching method for numerical data. The imputed data were obtained by calculating the mean value from 10 data sets and evaluated by checking the distribution (such as mean, quartiles, histogram) and the t-test of the observed and imputed numbers. All statistical analyses were performed in R version 3.6.2.¹⁶

Results

In the original data set, 461 women were recruited from the three hospitals and completed baseline interviews. After excluding women without hysterectomy details in their medical records (n = 9) and the women whose self-reported race was neither Black nor White (n = 21), there were 431 women included in the analyses.

Participant characteristics and hysterectomy approach counts by race

The demographic, baseline, and operative characteristics are presented for Black women and White women in Table 1. Compared with White women, Black women had higher BMI, were more likely to have public insurance, and had worse financial toxicity (lower) scores. Black women also tended to have greater EBL, longer procedure durations, and greater uterine weights. Black women were less likely to have had oophorectomy (ovary removal). In addition, compared with White women, Black women were more likely to have fibroids or have heavy bleeding and less likely to have endometriosis as an indication for surgery.

Table 1.

Characteristics of Women with All Planned Procedure Routes by Race Group in the Detroit Hysterectomy Study (Years 2015–2017) (n = 431)

Characteristic	White (n = 235)	Black (n = 196)	p ^a
Demographic
Age at verbal consent date (years), mean (SD)	45.60 (8.76)	46.56 (6.92)	0.213
BMI (kg/m²), mean (SD)^NA=18	32.25 (8.20)	33.94 (6.93)	<0.05^*
Insurance type, n (%)^NA=3			<0.05
Private/military	187 (79.6)	132 (68.4)
Public	48 (20.4)	61 (31.6)
Presurgery interview (baseline)
Current pain level, mean (SD)^NA=2 (0 = no pain, 10 = extreme pain)	2.93 (2.81)	2.68 (3.26)	0.395
PHQ-9 score, mean (SD) (highest score is associated with depression)^NA=6	5.53 (5.24)	5.89 (5.76)	0.494
Satisfaction,^b n (%)^NA=3	225 (96.2)	188 (96.9)	0.875
Financial toxicity (lower scores suggest worse financial toxicity), mean (SD)^NA=57	31.67 (8.92)	28.74 (9.25)	<0.05
Operative factors
Estimated blood loss (mL), mean (SD)^NA=6	143.6 (182.72)	196.30 (215.94)	<0.05
Postoperative transfusion, n (%)^NA=2	0 (0.0)	3 (1.5)	0.183
Any complications,^c n (%)^NA=1	15 (6.4)	14 (7.1)	0.913
Procedure duration (hours), mean (SD)	2.26 (0.95)	2.77 (1.21)	<0.05
Length of stay >1 day, n (%)	215 (91.5)	180 (91.8)	1
Uterine weight (g), mean (SD)^NA=2	243.68 (302.98)	495.75 (645.93)	<0.05
Had oophorectomy, n (%)	94 (40.0)	42 (21.4)	<0.05
Indication for surgery is fibroids, n (%)	116 (49.4)	164 (83.7)	<0.05
Indication for surgery is pelvic pain, n (%)	104 (44.3)	77 (39.3)	0.346
Indication for surgery is heavy bleeding, n (%)	131 (55.7)	134 (68.4)	<0.05
Indication for surgery is endometriosis, n (%)^NA=1	18 (7.7)	3 (1.5)	<0.05

The values in bold were statistically significant.

Unadjusted p-value: chi-squared test for categorical variables and t-test for continuous variables to test the difference between two race groups.

Satisfaction defined as reporting Strongly Agree or Agree to the statement: “I was satisfied with my decision.”

Any complications include the five intraoperative adverse events (converted to open, transfusion due to intraoperative blood loss, injury to “other” structures, other events, and death).

BMI, body mass index; n (%), count and column percent; NA, number of missing values; PHQ-9, Patient Health Questionnaire-9; SD, standard deviation.

Table 2 presents the frequencies of hysterectomies by planned approach and race group. Compared with White women, Black women had a lower proportion of planned MIS (65.3% vs. 80.4%, p < 0.05), planned abdominal laparoscopic (with or without robotic assist) (49% vs. 60%, p = 0.02), and planned vaginal (with or without laparoscopic) hysterectomies (16.3% vs. 20.4%, p = 0.33). Six women had a planned abdominal laparoscopic hysterectomy that was converted to an open surgery. Five women had a planned vaginal hysterectomy that was converted to an open surgery.

Table 2.

Planned Hysterectomy Procedure Route by Race Group

Procedure route	Overall (n = 431), n (%)	White (n = 235), n (%)	Black (n = 196), n (%)
Laparotomic/“open” abdominal hysterectomy (non-MIS)	114 (26.5)	46 (19.6)	68 (34.7)
MIS hysterectomy	317 (73.5)	189 (80.4)	128 (65.3)
(1) Abdominal laparoscopic (with/without robot assist)	237 (55.0)	141 (60.0)	96 (49.0)
(2) Vaginal (with/without laparoscopy)	80 (18.6)	48 (20.4)	32 (16.3)

MIS, minimally invasive surgery.

Demographic, clinical, and procedural details were compared between Black women and White women within each planned surgical approach group. Among the women who planned an “open” abdominal hysterectomy, Black women were more likely to have had public insurance compared with White women; however, this was not true for women who had planned MIS approaches (laparoscopic or vaginal) (Supplementary Tables S1–S3). Black women who had planned an abdominal or laparoscopic hysterectomy were less likely to have an oophorectomy compared with White women (Supplementary Tables S1 and S2). Black women who had a planned laparoscopic hysterectomy had worse financial toxicity (lower scores) than White women (Supplementary Table S2).

Regardless of the planned surgical approach, Black women tended to have longer procedure durations and larger uteri (Supplementary Tables S1–S3). Black women who had a planned MIS (laparoscopic or vaginal) but not a planned “open” hysterectomy were more likely than White women to have had a surgical indication of fibroids.

Demographic, clinical, and operative characteristics by planned procedure route

Race, EBL, uterine weight, proportion of women with a surgical indication of fibroids, and length of stay (>1 day) varied by planned procedure route (Table 3). In comparison with the women who had a planned “open” procedure, the women who had a planned MIS surgery, either abdominal laparoscopic procedure or any vaginal procedure, had lower EBL (abdominal laparoscopic vs. open: 108.9 ± 108.4 mL vs. 283.2 ± 259.9 mL [p < 0.05]; vaginal vs. open: 178.1 ± 238.2 mL vs. 283.2 ± 259.9 mL [p < 0.05]), lighter uterine weight (abdominal laparoscopic vs. open: 252.6 ± 251.5 g vs. 711.6 ± 806.2 g [p < 0.05]; vaginal vs. open: 165.1 ± 112.8 g vs. 711.6 ± 806.2 g [p < 0.05]), were less likely to have a surgical indication of fibroids (abdominal laparoscopic vs. open: 59.5% vs. 86.8% [p < 0.05]; vaginal vs. open: 50.0% vs. 86.8% [p < 0.05]), and were less likely to have a length of stay of more than 1 day (abdominal laparoscopic vs. open: 87.8% vs.100% [p < 0.05]; vaginal vs. open: 91.2% vs.100% [p < 0.05]).

Table 3.

Demographic, Presurgery Interview, Clinical, and Operative Characteristics by Planned Procedure Route

Characteristic	Total (n = 431)	Procedure route			p ^a
Characteristic	Total (n = 431)	Abdominal laparoscopic with/without robot assist (n = 237)	Vaginal with/without laparoscopy (n = 80)	“Open” abdominal (n = 114)	p ^a
Demographic
Race, n (%)					<0.05^*
White/Caucasian	235 (54.5)	141 (59.5)	48 (60.0)	46 (40.4)
African American	196 (45.5)	96 (40.5)	32 (40.0)	68 (59.6)
Age at verbal consent date (years), mean (SD)	46.04 (7.98)	45.56 (8.38)	45.63 (7.72)	47.30 (7.19)	0.141
BMI (kg/m²), mean (SD)^NA=18	33.04 (7.67)	33.55 (7.89)	31.82 (6.81)	32.85 (7.75)	0.219
Insurance type, n (%)^NA=3					0.144
Private/military	319 (74.5)	185 (78.1)	58 (72.5)	76 (68.5)
Public	109 (25.5)	52 (21.9)	22 (27.5)	35 (31.5)
Presurgery interview (baseline)
Current pain level, mean (SD)^NA=2 (0 = no pain, 10 = extreme pain)	2.82 (3.02)	2.85 (3.01)	3.14 (3.19)	2.53 (2.93)	0.377
PHQ-9 score, mean (SD) (highest score is associated with depression)^NA=6	5.69 (5.48)	5.33 (5.19)	6.28 (5.39)	6.02 (6.09)	0.318
Satisfaction,^b n (%)^NA=3	15 (3.5)	226 (96.2)	77 (97.5)	110 (96.5)	0.863
Financial toxicity (lower scores suggest worse financial toxicity), mean (SD)^NA=57	30.33 (9.18)	30.28 (9.42)	31.16 (9.26)	29.85 (8.64)	0.659
Operative factor
Estimated blood loss (mL), mean (SD)^NA=6	167.92 (200.22)	108.89 (108.41)	178.08 (238.20)	283.17 (259.91)	<0.05
Postoperative transfusion, n (%)^NA=2	3 (0.7)	1 (0.4)	0 (0.0)	2 (1.8)	0.267
Any complications,^c n (%)^NA=1	29 (6.7)	15 (6.4)	6 (7.5)	8 (7.0)	0.931
Procedure duration in hours, mean (SD)	2.49 (1.10)	2.74 (1.20)	2.09 (0.83)	2.26 (0.92)	<0.05
Length of stay >1 day, n (%)	395 (91.6)	208 (87.8)	73 (91.2)	114 (100.0)	<0.05
Uterine weight (g), mean (SD)^NA=2	358.26 (504.85)	252.58 (251.47)	165.12 (112.76)	711.63 (806.18)	<0.05
Had oophorectomy, n (%)	136 (31.6)	72 (30.4)	28 (35.0)	36 (31.6)	0.744
Indication for surgery is fibroids, n (%)	280 (65.0)	141 (59.5)	40 (50.0)	99 (86.8)	<0.05
Indication for surgery is pelvic pain, n (%)	181 (42.0)	99 (41.8)	40 (50.0)	42 (36.8)	0.187
Indication for surgery is heavy bleeding, n (%)	265 (61.5)	143 (60.3)	46 (57.5)	76 (66.7)	0.375
Indication for surgery is endometriosis, n (%)^NA=1	21 (4.9)	18 (7.6)	2 (2.5)	1 (0.9)	<0.05

The values in bold were statistically significant.

Chi-squared test for categorical variable and one-way ANOVA for continuous variable to test the difference among three procedure routes.

Satisfaction defined as reporting Strongly Agree or Agree to the statement: “I was satisfied with my decision.”

Any complications include the five intra-operative adverse events (converted to open, transfusion due to intra-operative blood loss, injury to ‘other’ structures, other events, and death).

ANOVA, analysis of variance; n (%), count and column percent; NA, number of missing values; SD, standard deviation.

Association between race and planned hysterectomy procedure route

Black women, compared with White women, were less likely to have had a planned MIS hysterectomy (abdominal laparoscopic or any vaginal, OR = 0.46, 95% CI: 0.3–0.71, p < 0.05), abdominal laparoscopic hysterectomy only (RRR = 0.46, 95% CI 0.29–0.73, p < 0.05), or a vaginal hysterectomy only (RRR = 0.45, 95% CI 0.25–0.81, p < 0.05) (Table 4). However, these racial differences did not remain statistically significant after adjusting for both confounders (Supplementary Table S4)—uterine weight and surgical indication of fibroids (MIS vs. abdominal [aOR = 0.93, 95% CI 0.55–1.57, p = 0.79], abdominal laparoscopic vs. open [aRRR = 0.89, 95% CI 0.52–1.51, p = 0.54], and vaginal vs. open [aRRR = 1.22, 95% CI 0.61–2.45, p = 0.58]) (Table 4).

Table 4.

Odds Ratios and Relative Risk Ratios of Black Women (vs. White Women) Planning a Minimally Invasive Surgery Approach Versus Abdominal Approach (Non-MIS)

	Model^a: MIS^b versus abdominal (non-MIS)		Model^c: abdominal laparoscopic (with/without robot assist) versus abdominal		Model^c: vaginal (with/without laparoscopy) versus abdominal
	Odds ratio (95% CI)	p	Relative risk ratio (95% CI)	p	Relative risk ratio (95% CI)	p
Race only (unadjusted)
Race: Black (Ref: White)	0.46 (0.3–0.71)	<0.05^*	0.46 (0.29–0.73)	<0.05	0.45 (0.25–0.81)	<0.05
Race + covariates^d (adjusted)
Race: Black (Ref: White)	0.93 (0.55–1.57)	0.79	0.89 (0.52–1.51)	0.54	1.22 (0.61–2.45)	0.58

The values in bold were statistically significant.

Logistic regression models for the outcome of MIS versus abdominal (non-MIS). The reference level is “Abdominal.”

MIS, minimally invasive surgery hysterectomy including laparoscopic (with/without robot assist) and vaginal (with/without laparoscopic) procedure routes.

Multinomial logistic regression models for the three procedure routes (abdominal, abdominal laparoscopic, and vaginal related) with the reference level of “Abdominal.”

The models adjusted for both uterine weight and surgery reason is fibroids.

CI, confidence interval.

Discussion

After adjusting for confounders, uterine weight and a surgical indication of fibroids, there was no difference in the likelihood of having planned an MIS hysterectomy and either laparoscopic or vaginal hysterectomy approaches between Black women and White women in a single health system in the Midwest. These results are consistent with some prior studies.^3

–8 Greater uterine weight and higher rate of surgical indication of fibroids were independently associated with lower odds of having MIS. In our data, Black women tended to have greater uterine weight, lower financial toxicity scores (worse financial toxicity), higher BMI, a higher proportion of public insurance, and higher proportions of surgical indication of fibroids and heavy bleeding. However, these variables, with the exception of uterine weight and surgical indication of fibroids, did not importantly change the associations between race and planned surgical approach. These results emphasize the need to consider all important confounders, including uterine weight and surgical indication, in analyses of racial differences in the likelihood of planning an MIS hysterectomy.

Some prior studies have been limited by the omission of key clinical details such as uterine weight or BMI, the lack of financial impact assessment, or insufficient sample size to permit comparisons across race groups—all concerns addressed in this study. In a study of all hysterectomies in the state of Maryland from July 2012 to September 2014 (n = 5,660), Black women were less likely (aOR = 0.7, 95% CI 0.63–0.78) than White women to have a minimally invasive hysterectomy (defined as laparoscopic hysterectomy, vaginal hysterectomy, or robot-assisted procedures) after adjusting for patient age, payer (insurance type), hospital bed number, surgeon and hospital hysterectomy volume, surgeon practicing year, Elixhauser Comorbidity Index, and surgical indication.³

Analyses of women who had TRICARE from 2006 to 2010 (n = 33,015), universal insurance coverage for U.S. Uniformed Service members, military retirees, and their dependents, also found that Black women were less likely than White women to have either a total vaginal hysterectomy (aRRR = 0.63, 95% CI 0.58–0.69) or total laparoscopic hysterectomy (aRRR = 0.65, 95% CI 0.60–0.71) after adjustment for various potential confounders, including age, marital status, region, system of care, beneficiary category, and sponsor rank (a proxy for socioeconomic status).⁸ Nearly 40% of women were excluded from the TRICARE analyses due to missing race data. The Maryland and TRICARE analyses relied on claims data, and no data were available on clinical characteristics such as uterine weight—which was an important confounder in our analyses and its omission could explain the difference in results between these and our study. Additionally, we used planned surgical approach rather than completed procedures; however, only 11 women in this study had a planned MIS approach converted to an open surgery.

In a study of hysterectomy patients (n = 1,746) from three urban “university-affiliated” hospitals in Philadelphia, PA, unadjusted analyses indicated that African American women were less likely to have any type of minimally invasive hysterectomy (OR = 0.80, 95% CI 0.65–0.97).⁷ However, in analyses adjusted for age, BMI, uterine weight, income quartile, payer (insurance type), and other potential confounders (obstetrical and surgical history, procedure year, and surgical indication), the authors suggest that this disparity was ameliorated (aOR = 0.82, 95% CI 0.61–1.10) since the aOR was not statistically significant. The OR and the aOR only changed by ∼2.5%, suggesting that all the factors included in the adjusted analyses did not actually confound the race–surgical approach relationship. Adjusting for all these extraneous factors in the model likely reduced the precision of the estimated OR and this could have led to the change in statistical significance. In further contrast with our analyses, we used an individual-level measure of financial toxicity, whereas the Philadelphia study used the estimated median household income (granularity level to zip code).⁷

Results from analyses of the 2015 National Surgical Quality Improvement Program's (NSQIP) “targeted hysterectomy file” also found that Black women were more likely to have a hysterectomy with an open approach compared with White women (aOR = 2.02, 95% CI 1.85–2.20).⁵ This OR was adjusted for uterine weight, prior abdominal surgery, pelvic surgery, endometriosis, BMI, age, and Charlson Comorbidity Index. The analyses of the NSQIP data included important patient-level factors (uterine weight, obesity, and comorbidities). However, our work was conducted at a single health system where all eligible patients, in theory, had access to MIS and this may not be true of all patients in the NSQIP database, which suggests that patient location is likely an important contributor to the reported racial difference.

This study has limitations that should be noted. The data source we used is limited to a single large health system, but this system consists of multiple hospital facilities and serves a socioeconomically and racially diverse population. We did not ask women why they selected their specific provider. Also, we did not collect participants' surgical history or parity. Additionally, this study did not capture participant history of prior vaginal delivery; however, it has been shown that 92% of vaginal hysterectomies planned for a cohort of women with no prior vaginal deliveries could be successfully completed with a vaginal approach.¹⁷

While HFHS has a cadre of physicians that are trained and experienced in the performance of MIS hysterectomy, all HFHS gynecologists do not perform MIS hysterectomy. Within the HFHS medical group employed physician staff model, patients can be referred to another provider at HFHS who can perform an MIS hysterectomy for them; however, the women may prefer to stay with their doctor with whom they may have a long-term relationship or their doctor may suggest that as their long-term provider, they are ideally suited to provide the best care for the woman because of their detailed knowledge of the woman's health, circumstances, and the reasoning behind her prior care choices.

In this study, we did not collect data about which women initially sought care from an MIS specialist or were referred to such as surgeon. In focus groups with prior hysterectomy patients, it was not uncommon for women to report that they did not discuss surgical approaches with their doctor because they trusted their doctor and were comfortable following their doctor's recommendations.¹⁸ We did not collect information on the surgical approaches offered to each woman and why they were offered. Therefore, we suggest that it is possible that women may appear to have a choice, but the options in that choice may not be perceived as equally acceptable to the woman in her decision-making process.

Conclusions

In this study at a single health system in the Midwest, Black women were not less likely than White women to have an MIS hysterectomy after controlling for important clinical variables (uterine weight and surgical indication of fibroids). These data provide evidence of the importance of these factors in explaining racial differences and should be considered in all future analyses of racial differences in MIS surgery rates among Black women and White women. Treatment decision-making at the intersection of patient and physician, including understanding actual availability of treatment options, remains understudied and could likely contribute key knowledge in understanding observed disparities.

Footnotes

Authors' Contributions

Study design: G.W. Data acquisition or analysis: W-.T.K.S., C.M.C., A.S.B., and G.W. Article drafting and approval of submitted article: all authors. Critical revision: W-.T.K.S. and G.W.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This study was funded by the Agency for Healthcare Research and Quality (Grant No. R24 HS022417).

Supplementary Material

Supplementary Table S1

Supplementary Table S2

Supplementary Table S3

Supplementary Table S4

References

Wilcox

, Koonin

, Pokras

, et al. Hysterectomy in the United States, 1988–1990. Obstet Gynecol, 1994; 83:549–555.

Louie

, Strassle

, Moulder

, et al. Uterine weight and complications after abdominal, laparoscopic, and vaginal hysterectomy. Am J Obstet Gynecol, 2018; 219:480.e1–480.e8.

Mehta

, Xu

, Hutfless

, et al. Patient, surgeon, and hospital disparities associated with benign hysterectomy approach and perioperative complications. Am J Obstet Gynecol, 2017; 216:497.e1–497.e10.

Abenhaim

, Azziz

, Hu

, et al. Socioeconomic and racial predictors of undergoing laparoscopic hysterectomy for selected benign diseases: Analysis of 341 487 hysterectomies. J Minim Invasive Gynecol, 2008; 15:11–15.

Alexander

, Strohl

, Rieder

, et al. Examining disparities in route of surgery and postoperative complications in black race and hysterectomy. Obstet Gynecol, 2019; 133:6.

Cohen

, Vitonis

, Einarsson

. Updated hysterectomy surveillance and factors associated with minimally invasive hysterectomy. JSLS, 2014; 18:e2014.00096.

Price

, Zimmerman

, Koelper

, et al. Social determinants of access to minimally invasive hysterectomy: Reevaluating the relationship between race and route of hysterectomy for benign disease. Am J Obstet Gynecol, 2017; 217:572.e1–572.e10.

Ranjit

, Sharma

, Romano

, et al. Does universal insurance mitigate racial differences in minimally invasive hysterectomy?. J Minim Invasive Gynecol, 2017; 24:790–796.

ACOG. ACOG Committee Opinion No. 649: Racial and ethnic disparities in obstetrics and gynecology. Obstet Gynecol, 2015; 126:e130–e134.

10.

Kassem

, Coleman

, Bossick

, et al. Patient perceptions of planned organ removal during hysterectomy. J Patient Cent Res Rev, 2019; 6:28.

11.

Sangha

, Bossick

, Su

W-TK

, et al. A prospective study of patterns of regret in the year after hysterectomy. J Patient Cent Res Rev, 2020; 7:329.

12.

Kroenke

, Spitzer

, Williams

. The PHQ-9: Validity of a brief depression severity measure. J Gen Intern Med, 2001; 16:606–613.

13.

De Souza

, Yap

, Wroblewski

, et al. Measuring financial toxicity as a clinically relevant patient-reported outcome: The validation of the COmprehensive Score for financial Toxicity (COST). Cancer, 2017; 123:476–484.

14.

O'Connor

. Validation of a decisional conflict scale. Med Decis Making, 1995; 15:25–30.

15.

LaMorte

. Confounding and effect measure modification. Available at: https://sphweb.bumc.bu.edu/otlt/MPH-Modules/BS/BS704-EP713_Confounding-EM/BS704-EP713_Confounding-EM_print.html Accessed June 5, 2021 .

16.

Buuren

, Groothuis-Oudshoorn

. Mice: Multivariate imputation by chained equations in R. J Stat Softw, 2010; 45:1–68.

17.

Le Tohic

, Dhainaut

, Yazbeck

, et al. Hysterectomy for benign uterine pathology among women without previous vaginal delivery. Obstet Gynecol, 2008; 111:829–837.

18.

Bossick

, Sangha

, Olden

, et al. Identifying what matters to hysterectomy patients: Postsurgery perceptions, beliefs, and experiences. J Patient Cent Res Rev, 2018; 5:167–175.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB

0.01 MB