Kenkyu |Villarreal MD | Mobily ME | Vasan S | Kalesan B |The Influence of Neighborhood | Income | Level on Utilization of | Laparoscopy | for Colorectal Procedures | Laparoscopy | Proctectomy | Colectomy | Colorectal | Income | Neighborhood | Affluence | Utilization | Disparity |Laparoscopy | Proctectomy | Colectomy | Colorectal | Income | Neighborhood | Affluence

Abstract

Background

While the field of healthcare has placed a strong emphasis on patient-centered medicine, it remains important to consider distal epidemiologic determinants of patient care. Determinants of health have an effect on not only patient outcomes, but also influence clinical behavior and financial consequences. We examined the Nationwide Inpatient Sample from 2009-2011 for the effect of neighborhood income level on whether patients undergoing colectomy or proctectomy received an open or laparoscopic procedure.

Methods

Laparoscopic and open colectomy/proctectomy procedures were identified by International Classification of Diseases-9 codes. The exposure was neighborhood income level, and outcomes were utilization of laparoscopic and open colorectal procedures, and total hospital charges.

Results

One-third of patients undergoing colorectal procedures underwent laparoscopic open surgery. After adjustment for covariates, compared to the 1st (poorest) quartile, the likelihood of undergoing laparoscopic colectomy/proctectomy was significantly higher in the 2nd (OR=1.14, 95%-CI:1.08-1.22), 3rd (OR=1.25, 95%-CI:1.16-1.36) and 4th (OR=1.49, 95%-CI:1.34-1.65) quartiles. This effect was modified by gender, age, mortality risk, and cancer diagnosis (p-interaction=<0.0001, <0.0001, 0.0029, <0.0001, respectively). There was a reduction in hospital charges associated with laparoscopic compared to open procedures, which improved with increasing patient income (change: -$1,518.40, p-trend<0.0001).

Conclusions

Patients from high-income neighborhoods preferentially received minimally invasive colorectal procedures. This utilization was greater among men, younger adults, those with little-to-no risk of mortality and those who were free of cancer at hospitalization. The cost savings were most pronounced among those from high-income neighborhoods.

Keywords: Laparoscopy; Proctectomy; Colectomy; Colorectal; Income; Neighborhood; Affluence; Utilization; Disparity.

Introduction

The use of laparoscopy for colectomy and proctectomy has increased in the past few years [1]. While an open procedure (OP) is at times the best option for patients, laparoscopic procedures (LP) are often performed [2]. Adoption of OP has been variable and traditionally underutilized, with about 7.3% of colorectal surgical patients undergoing LP at high-volume centers [3] [4]. The more recent trend of LP adoption can be attributed to decreased length of hospital stay, decreased complications, and a quicker return to normal activities [5] [6] [7] [8]. Various barriers to health care and access have stymied the universal adoption of the minimally invasive approach nationally [9]. Some of these barriers at the patient, provider, and system level include social structure, living conditions, lifestyle, family support, financial means, culture, language skills, health beliefs, values concerning illness, surgeon training, and transportation [10] [11].

2. Literature Search

Laparoscopic techniques were applied to increasing numbers of general surgical procedures since the early 1990s [12]. Although this minimally invasive approach for colectomy was found to reduce hospitalization time, hasten return to normal activity, and reduce health care costs, the controversies revolved around a lack of prospective studies to validate superiority to open colectomy [13]. In the 21st century, there was evidence from clinical trials that established the effectiveness of laparoscopic colectomy over OP for treatment of colon cancer in terms of morbidity, hospital stay, tumor recurrence, and cancer-related survival [14-16].

However, recent work found a lack of clinical comparability between patients undergoing open and minimally invasive colorectal procedures; younger surgeons were found to perform LP at the expense of training for OP, while the patients at higher risk, indicated for OP, were receiving inadequate care [17]. There can also be surgeon and high volume center preferences towards laparoscopic procedures [18]. Outside of hospitals, differential utilization may be due to economic status [9], type of insurance [19], but not race [19] [20]. These utilization disparities can all be encompassed by the differences seen in economic status of each individual colorectal patient. While clinical features are important covariates to study, they might not entirely explain the primary reason that patients are originally presenting with high vs. low medical risk. While there currently is a strong emphasis on patient-centered medicine, it remains important for the field of public health to evaluate distal epidemiologic determinants of patient care in their causal fields of interest [21] [22].

Given the evidence on structural barriers influencing medical treatment, we decided to use the ecological model of health as the basis for our question [23]. The ecological model of health describes concentric realms of influence that change a patient’s outcome, from individual to interpersonal to community and up to societal characteristics [24]. In addition to accounting for usual disparities in health (e.g. age, gender, and ethnicity), the patient’s zip code reflects socioeconomic status and the level care available [25] [26]. Neighborhood median income can serve as a proxy for affluence while encompassing a large number of natural confounders [27]. We hypothesized that higher neighborhood income drives greater utilization of colorectal LP over OP. We also sought to

determine the differences by gender, age, risk of mortality, and cancer status; and then assessed the cost savings associated with the increased utilization of LP across neighborhood income quartiles.

Methods

We used cross-sectional nationally representative hospitalization data from Nationwide Inpatient Sample (NIS) from 2009 – 2011, which contains all-payer data on inpatient hospitalizations from participating states [28]. NIS is managed by the Healthcare Cost and Utilization Project (HCUP) and sponsored by the Agency for Healthcare Research and Quality.

The International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) procedure and diagnosis codes were used to identify all billed hospitalizations as primary colorectal procedures (colectomy or proctectomy). Procedures were classified as either “Open” or “Laparoscopic” based on billing codes. Previous studies have also included codes for LP converted to OP (V64.41), but that was omitted here because there has been an overall decrease in conversions due to better identification of patients for LP, surgeon experience, and evolution of the technology [29]. In addition, we kept the analysis intention-to-treat, in accordance with previous work [30].

Procedure	Code
ICD-9 Open
Open and other multiple segmental resection of large intestine	4571
Open and other cecectomy	4572
Open and other right hemicolectomy	4573
Open and other resection of transverse colon	4574
Open and other left hemicolectomy	4575
Open and other sigmoidectomy	4576
Other and unspecified partial excision of large intestine	4579
Open total intra-abdominal colectomy	4582
Open pull-through resection of rectum	4843
Open abdominoperineal resection of the rectum	4852
ICD-9 Lap
Laparoscopic total intra-abdominal colectomy	4581
Laparoscopic pull-through resection of the rectum	4842
Laparoscopic abdominoperineal resection of the rectum	4851
Laparoscopic multiple segmental resection of the large intestine	1731
Laparoscopic cecectomy	1732
Laparoscopic right hemicolectomy	1733
Laparoscopic resection of transverse colon	1734
Laparoscopic left hemicolectomy	1735
Laparoscopic sigmoidectomy	1736
Other laparoscopic partial excision of large intestine	1739

ICD-9-CM Procedure Codes for Laparoscopic and Open Colectomies and Proctectomies

4. Data Source

Data was obtained from the NIS Inpatient Core File, and is derived from approximately 8 million hospitalizations across 1,000 different hospitals annually, creating a 20% stratified probability sample of US hospitals [28]. We used NIS data from 2009 to 2011. The main hospitalization data were linked to NIS severity file data and NIS hospital data file elements.

5. Study Population and Variables

From all hospitalizations in the NIS data from 2009–2011, those hospitalizations with either colectomy or proctectomy indicated as the primary procedure were retained. The types of surgical procedures, laparoscopic and open identified using ICD-9 codes were the outcomes. The primary outcome measured was either an open or laparoscopic procedure. In 2011, there were 94,232 LP and 169,952 OP, in 2010 there were 79,095 and 168,837, while in 2009 it was 78,279 and 176,833, respectively. The main exposure was neighborhood income level, assessed using median income associated with zip code of each hospitalization. Patient neighborhood income level by assigning median income from that zip code to four quartiles based on US median incomes (1-38,999; 39,000-47,999; 48,000-63,999; 64,000+). Hospitalizations without zip code information (1.76% in 2011, 2.05% in 2010 and 2.18% in 2009) were excluded. The patient covariates used for analysis were age, gender, race, site of surgery (rectum versus colon), type of diagnosis (cancer versus no cancer), Diagnosis Related Groups (DRG) mortality risk, DRG severity and expected payer. Age was categorized into 5 groups: ≤45, 46-55, 56-65, 66-75 and >75 years. Race was missing for 11.4% of hospitalizations in the entire cohort. The hospital characteristics considered were based on teaching status, bed size, location, discharge percentiles, ownership, region, and total charges.

6. Statistical analysis

We used survey analysis methods as provided by HCUP survey weights. All covariates in this analysis were considered as categorical variables. A multivariable survey logistic regression model was used to determine the relationship between patient and hospital characteristics, and derive odds ratios (OR) and 95% confidence intervals (95%-CI) for LP versus OP. The lowest quartile for neighborhood income level was used as the reference group for OR and 95%-CI. The probability of LP and OP were predicted from the multivariable regression model. Stratified analysis by gender, age, cancer diagnosis and DRG mortality risk was performed using appropriate interaction tests. Predicted means and difference in predicted means of total charges between LP versus OP was obtained using a multivariable survey regression model. The variation by gender, age, cancer diagnosis and DRG mortality risk was also performed using stratified analysis. All analysis was done using STATA 13.1 (STATA Corp LP, College Station, TX, USA). All p-values and 95%-CI were 2-sided. We considered p<0.0001 to be statistically significant due to the large sample size [31].

Results

Baseline characteristics of all patients receiving OP and LP are presented and compared in Table 1. In total, 32.8% were LP and 67.2% were OP. Out of all colorectal surgery hospitalizations from 2009-2011, 19.0% of LP were in the lowest income quartile, 23.7% were in quartile 2, 27.2% were in the third quartile, and 30.1% were in the highest quartile. 26.2% of OP were in the lowest, 26.5% were in 2nd quartile, 25.5% were in quartile 3, and 21.8% were in the highest group. When compared to the lowest neighborhood income quartile group, utilization of LP versus OP, after adjusting for covariates, increased 14% for the second quartile (OR=1.14, 95%-CI:1.08-1.22), 25% for the third (OR=1.25, 95%-CI:1.16-1.36), and 49% for patients from the wealthiest neighborhood quartile (OR=1.49, 95%-CI:1.34-1.65). In the multivariable regression, neighborhood income level, age, race, surgery site, cancer diagnosis, hospital location, DRG mortality, DRG severity and expected payer were independently associated with type of surgical procedure.

	Lap, %	Open, %	Crude OR (95% CI)	Crude p	Adj OR (95% CI)	Adj p
Patient neighborhood affluence				<0.0001		<0.0001
1^st quartile (lowest)	19.0	26.2	Reference		Reference
2^nd quartile	23.7	26.5	1.23 (1.16-1.30)		1.14 (1.08-1.22)
3^rd quartile	27.2	25.5	1.47 (1.37-1.57)		1.25 (1.16-1.36)
4^th quartile (highest)	30.1	21.8	1.90 (1.75-2.06)		1.49 (1.34-1.65)
Age				<0.0001		0.0001
<=45	14.6	14.0	Reference		Reference
46-55	19.3	15.8	1.17 (1.12-1.22)		0.97 (0.92-1.02)
56-65	24.0	21.2	1.09 (1.04-1.14)		0.94 (0.89-0.99)
66-75	23.0	22.7	0.97 (0.92-1.02)		1.06 (0.99-1.13)
76+	19.1	26.2	0.70 (0.66-0.74)		1.05 (0.97-1.13)
Gender				0.59		0.54
Male	46.7	46.6	Reference		Reference
Female	53.3	53.4	0.99 (0.97-1.02)		0.99 (0.97-1.02)
Race				<0.0001		<0.0001
White	79.7	77.3	Reference		Reference
Black	8.5	10.8	0.76 (0.70-0.82)		0.81 (0.75-0.88)
Hispanic	7.1	7.2	0.96 (0.81-1.14)		1.03 (0.88-1.20)
Other	4.7	4.6	0.98 (0.88-1.10)		0.98 (0.88-1.08)
By site				<0.0001		<0.0001
Rectum	2.2	2.8	Reference		Reference
Colon	97.8	97.2	1.28 (1.14-1.44)		1.72 (1.52-1.96)
By diagnosis				<0.0001		<0.0001
Non-cancer	41.0	54.5	Reference		Reference
Cancer	59.0	45.5	1.73 (1.66-1.80)		1.73 (1.66-1.81)
Hospital teaching status				0.001		0.78
Non-teaching	50.1	54.4	Reference		Reference
Teaching	49.9	45.6	1.19 (1.07-1.32)		0.98 (0.85-1.13)
Bed size of hospital				0.0063		0.0361
Small	9.9	11.9	Reference		Reference
Medium	24.6	23.6	1.25 (1.08-1.45)		1.22 (1.03-1.45)
Large	65.4	64.5	1.21 (1.06-1.39)		1.08 (0.88-1.31)
Hospital location				<0.0001		<0.0001
Rural	7.2	13.1	Reference		Reference
Urban	92.8	86.9	1.96 (1.69-2.27)		1.71 (1.41-2.08)
Total number of discharges				<0.0001		0.0242
<=25^th percentile	21.1	27.3	Reference		Reference
25^th – 50^th percentile	25.1	25.1	1.29 (1.13-1.48)		1.15 (0.98-1.35)
50^th – 75^th percentile	25.0	24.7	1.31 (1.15-1.49)		1.13 (0.92-1.40)
>75^th percentile	28.8	22.9	1.63 (1.41-1.89)		1.43 (1,11-1.83)
Hospital ownership				0.0034		0.0358
Government/ non-federal	9.3	11.2	Reference		Reference
Private, non-profit	80.1	76.7	1.26 (1.06-1.51)		1.24 (1.01-1.51)
Private, invest-own	10.6	12.1	1.06 (0.86-1.31)		1.05 (0.84-1.33)
Region of hospital				0.087		0.0041
Northeast	19.6	19.2	Reference		Reference
Midwest	22.4	25.0	0.88 (0.74-1.03)		1.12 (0.93-1.35)
South	39.8	37.2	1.05 (0.91-1.21)		1.30 (1.11-1.52)
West	18.2	18.6	0.96 (0.81-1.14)		1.05 (0.87-1.27)
DRG mortality risk				<0.0001		<0.0001
No class	0.02	0.01	Reference		Reference
Minor likelihood of dying	64.8	35.9	Reference		Reference
Moderate likelihood of dying	23.7	26.6	0.50 (0.48-0.51)		0.62 (0.60-0.65)
Major likelihood of dying	8.1	20.4	0.22 (0.21-0.23)		0.42 (0.39-0.44)
Extreme likelihood of dying	3.7	17.1	0.11 (0.10-0.12)		0.32 (0.29-0.36)
DRG severity				<0.0001		<0.0001
No class	0.02	0.01	Reference		Reference
Minor loss of function	44.1	17.7	Reference		Reference
Moderate loss of function	37.5	33.1	0.45 (0.44-0.47)		0.58 (0.56-0.60)
Major loss of function	14.1	27.9	0.20 (0.19-0.21)		0.37 (0.35-0.40)
Extreme loss of function	4.2	21.3	0.08 (0.07-0.08)		0.24 (0.22-0.27)
Expected payer				<0.0001		<0.0001
Medicare	42.3	50.8	Reference		Reference
Medicaid	3.9	7.3	0.65 (0.60-0.70)		0.66 (0.61-0.72)
Private, HMO	49.2	34.9	1.69 (1.63-1.75)		1.18 (1.13-1.24)
Self-pay	1.9	4.0	0.57 (0.51-0.63)		0.47 (0.42-0.54)
No charge/ other	2.6	3.1	1.00 (0.89-1.11)		0.77 (0.68-0.87)

Table 1: Association of laparoscopic and open colectomy utilization and patient neighborhood affluence.

All percentages are survey weighted. Lap: Laparoscopic; OR: Odds Ratio; CI= confidence intervals
Crude and adjusted OR and 95% CI are derived from survey logistic regression. Race missing for 17,730 hospitalizations.
The horizontal line indicates predicted probabilities from multivariate survey logistic regression and the vertical lines indicate 95% confidence interval for probability estimates.
Figure 1 shows the predicted probabilities of colectomy and proctectomy procedures from 2009 to 2011. In the lowest income quartile (<$39,000), 30% of all procedures were laparoscopic and there was a gradual increase in LP with increasing median household income to 38% in the highest group. Multivariable survey logistic regression showed a decrease in the rates of OP from 70% to 62% as the median household income quartiles increase.

Figure 1: Predicted probability of receiving laparoscopic and open colorectal procedure within each income quartile.

The difference by gender in the utilization of LP over OP and neighborhood income level is represented in Figure 2a and Supplementary Table 2. Overall, men were more likely than women to receive LP than OP and this trend was more pronounced with increased neighborhood income level (p-interaction<0.0001). In comparison to those in the poorest neighborhoods, men in the second lowest income level quartile were more likely to utilize LP (OR=1.24, 95%-CI:1.16-1.34) and this trend continued in the second highest category (OR=1.47, 95%-CI:1.35-1.59) and the highest income level (OR=1.93, 95%-CI:1.75-2.12). Women also showed a higher probability of utilizing LP from the second quartile (OR=1.19, 95%-CI:1.11-1.27), the third quartile (OR=1.41, 95%-CI:1.31-1.52), and the highest (OR=1.72, 95%-CI:1.57-1.88).

Figure 2a: The horizontal line indicates predicted probabilities from multivariate survey logistic regression and the vertical lines indicate 95% confidence interval for probability estimates.

Figure 2b: The horizontal line indicates predicted probabilities from multivariate survey logistic regression and the vertical lines indicate 95% confidence interval for probability estimates.

	Patient neighborhood affluence
	39,000 to 47,999	48,000 to 63,999	64,000+	p	p-inter

Gender					<0.0001
Men	1.24 (1.16-1.34)	1.47 (1.35-1.59)	1.93 (1.75-2.12)	<0.0001
Women	1.19 (1.11-1.27)	1.41 (1.31-1.52)	1.72 (1.57-1.88)	<0.0001
Age					<0.0001
<=55	1.21 (1.12-1.31)	1.49 (1.37-1.62)	2.05 (1.87-2.26)	<0.0001
56-65	1.26 (1.16-1.38)	1.48 (1.35-1.63)	1.96 (1.75-2.18)	<0.0001
66-75	1.20 (1.10-1.30)	1.43 (1.30-1.57)	1.69 (1.51-1.89)	<0.0001
76+	1.19 (1.09-1.31)	1.33 (1.20-1.47)	1.51 (1.33-1.70)	<0.0001
Diagnosis of cancer					<0.0001
No	1.25 (1.17-1.35)	1.53 (1.40-1.66)	2.08 (1.87-2.31)	<0.0001
Yes	1.21 (1.13-1.30)	1.43 (1.32-1.54)	1.71 (1.56-1.88)	<0.0001
DRG risk of mortality					0.0029
None, minor, moderate	1.20 (1.13-1.28)	1.43 (1.33-1.54)	1.82 (1.66-2.00)	<0.0001
Major, extreme	1.17 (1.05-1.29)	1.31 (1.18-1.46)	1.52 (1.34-1.71)	<0.0001

Supplementary Table 2: Stratified analysis of utilization of surgical procedures and patient neighborhood affluence

Adjusted for age, sex and race. Income <39,000 (first quartile) serves as reference

Figure 2b and Supplementary Table 2 demonstrate the effect of patient neighborhood affluence on utilization, stratified by age. Overall, younger patients were more likely than older patients to receive LP than OP among all levels of neighborhood affluence, and this difference was more pronounced with increasing neighborhood income level (p-interaction<0.0001). In the wealthiest compared to the poorest neighborhoods, utilization of LP over OP increased two-fold among patients ≤55 (OR=2.05, 95%-CI:1.87-2.26), 96% for those 56 to 65 (OR=1.96, 95%-CI:1.75-2.18), 69% for those 66 to 75 (OR=1.69, 95%-CI:1.51-1.89), and only 51% for patients 76 and older (OR=1.51, 95%-CI:1.33-1.70).

The difference by cancer diagnosis in the utilization of LP over OP and patient neighborhood affluence is represented in Figure 2c and Supplementary Table 2. Patients without cancer were more likely than those with cancer to receive laparoscopic than OP and this difference was more substantial with increased neighborhood affluence (p-interaction<0.0001). The likelihood for receiving laparoscopic over OP among the non-cancer hospitalizations was 25% higher in the 2nd quartile (OR=1.25, 95%-CI:1.17-1.35), 53% in the 3rd quartile (OR=1.53, 95%-CI:1.40-1.66) and greater than 2-fold higher in the 4th quartile (OR=2.08, 95%-CI:1.87-2.31) as compared to the lowest neighborhood affluence quartile. The likelihood for receiving laparoscopic over OP among the cancer hospitalizations were 21% more likely in the 2nd quartile (OR=1.21, 95%-CI:1.13-1.30), 43% in the 3rd quartile (OR=1.43, 95%-CI:1.32-1.54) and 71% higher in the 4th quartile (OR=1.71, 95%-CI:1.56-1.88) as compared to the lowest neighborhood affluence quartile.

Figure 2C: The horizontal line indicates predicted probabilities from multivariate survey logistic regression and the vertical lines indicate 95% confidence interval for probability estimates.

The association between neighborhood affluence and utilization also varied by DRG risk of mortality, between ‘little to no risk’ (None, minor, moderate) and ‘high risk’ (major, extreme) as shown in Figure 2d and Supplementary Table 2. Even though the higher utilization of LP over OP increased with greater neighborhood affluence among high and little to no DRG mortality risk, the magnitude of increase was greater among those with little to no risk of mortality (p-interaction=0.0029). For those with little to no risk, the utilization of LP over OP increased 20% for the second income quartile (OR=1.20, 95%-CI:1.13-1.28), 43% for the third (OR=1.43, 95%-CI:1.33-1.54) and 82% for the fourth (OR=1.82, 95%-CI:1.66-2.00). Those with high risk showed an increase of 17% for the second income quartile (OR=1.17, 95%-CI:1.05-1.29) to 31% in the third (OR=1.31, 95%-CI:1.18-1.46) and to 52% in the fourth (OR=1.52, 95%-CI:1.34-1.71) as compared to the lowest affluence quartile.

Figure 2D: The horizontal line indicates predicted probabilities from multivariate survey logistic regression and the vertical lines indicate 95% confidence interval for probability estimates.

Figure 3 summarizes the difference in total charges between LP versus OP, first according to neighborhood affluence quartiles, and then stratified by gender, age, cancer diagnosis and DRG mortality risk. Charges for OP were significantly higher than LP within each strata of neighborhood affluence; the differences were $33,978, $32,965, $37,993, and $36,205 in the 1st, 2nd, 3rd, and 4th quartile respectively (p<0.0001). With every quartile increase in neighborhood affluence, an increase of $1,518 (95%-CI:$932-$2,104) difference of total charges between laparoscopic and OP was observed. The reduction in hospital costs associated with LP over OP across increasing neighborhood affluence was significantly different between men and women ($1,435 and $1,633, p-interaction<0.0001, respectively). Similar significant differences were also found by age (<=55:$1,406, p-interaction<0.0001), cancer diagnosis (Yes:$1518, p-interaction<0.0001) and DRG risk of mortality (None to moderate: $2323, p-interaction<0.0001).

Figure 3: Total charges by utilization of surgical procedures according to quartiles of patient neighborhood affluence

Discussion

Using a large nationally representative hospitalization dataset from 2009 to 2011, we found that one-third of colectomies and proctectomies were performed laparoscopically. However, the utilization of LP increased with increasing patient neighborhood income level. Men, younger individuals, cancer free patients, and those with little to no risk of mortality by DRG were more likely to undergo the laparoscopic over open approach. Those undergoing LPs were also found to be associated with lower hospital costs than OPs, and this reduction increased with increasing neighborhood income level.

Our main finding of greater utilization of LP with an increase in neighborhood income level was also similar to the results of previous studies that suggested colorectal cancer patients whose income was above $39,000 had a greater chance of undergoing a LP compared to patients with a lower income [19]. The benefits of living in high-income neighborhoods were further explained via greater access to health care facilities and maintenance health care procedures (e.g. colonoscopies), that allowed for earlier detection and easier resection [19]. Another study on the effect of neighborhood on cardiac rehabilitation found lower rates of attendance and completion of cardiac rehabilitation among those who lived in low-income neighborhoods, regardless of increasing access [32]. This is troubling because it suggests that utilization of minimally invasive procedures is influenced by income or the paying capacity of the patient. Similarly, our main finding may be explained by income disparity; lower income categories tend to be homogenous due to the floor effect with clustering of incomes not greater than $65,000, [33] while among the wealthiest quartile, income ranges above $65,000 have no ceiling effect [34].

The effect of affluence on utilization of laparoscopic versus open surgery was modified by several factors. We found that men were more likely than women to utilize LP over OP, and this difference increased with increasing neighborhood income. Significant gender differences in utilization of health-care services were reported in a prospective study of older adults, even after adjusting for economic factors [35]. Another cohort study of men between 15-64 years demonstrated social class and socio-economic status (SES) to be strong drivers of preventable mortality; the risk of death was found to be higher in individuals of lower SES [36]. Taken together, these studies show that men may use certain surgical services more than women [35], and that this health seeking behavior is more pronounced for wealthier men [36], as illustrated in our results.

The effect of age on utilization of minimally invasive procedures according to neighborhood income level showed that among the most affluent quartile, younger patients were more likely to receive LP than older patients. Our results were in line with a report using 2007 NIS data which found that among those who had colon cancer, hospitalizations <70 years were more likely to receive laparoscopic resection as compared to open resection [30]. A study from the Netherlands found that elderly patients benefit the most in terms of absolute mortality risk reduction, but also showed that the laparoscopic approach was used less frequently than for younger patients [37].

Our study also found a differential in the relationship of neighborhood income level and procedure utilization by cancer diagnosis. Our study found greater utilization of a LP over OP among those with low risk and cancer-free as compared to the high risk and cancer patients. In 2012, the National Comprehensive Cancer Network guidelines for rectal cancer treatment only recommended laparoscopy within a study protocol and in a specialized center [38] [39]. These specialized centers may be more readily available to patients in the most affluent quartiles, increasing their likelihood of receiving a LP for rectal cancer [40]. With respect to colon cancer, randomized clinical trials (RCT’s) have shown that long-term oncogenic outcomes are comparable between laparoscopic and open colectomy [16,41-42].

We found that there was increased utilization in LP over OP in the lower DRG mortality risk group. This increase was greater among more affluent neighborhoods. One possible explanation is that DRG mortality risk is a proxy for cancer stage, as the correlation between increasing DRG scores and cancer metastasis has been previously shown [43]. Patients in higher income bracket and payer status were more likely to have a lower stage and less complicated colorectal cancer [44]. Therefore, higher affluence may be associated with a lower stage of cancer, i.e. lower DRG mortality risk, and related to a larger increase of laparoscopic utilization. However, our study was underpowered for examining a three-way interaction.

Our study showed large overall and subgroup-specific differences between the two procedures in terms of total hospital charges, with LP favoring lower costs. These differences have been suggested by previous studies and appears to be based mostly on postoperative outcomes [45][46]. Interestingly, increased neighborhood affluence augmented these cost savings. For each increase in neighborhood affluence quartile there was a significant increase in the cost savings of a laparoscopic compared to OP. These differences between different subgroups force us to speculate on the directionality; whether type of procedure drives the reduction in cost or whether the choice of surgical procedure is driven by the intention of cost reduction. In a study comparing laparoscopic to open colectomy (2003 and 2004 NIS), there was no difference in charges by type of procedure [30]. However, this study was performed at a time when laparoscopic colectomy had not yet been widely adopted; there were only 3,296 laparoscopic hospitalizations (3.3%) at the time of the study compared to 78,280 (32.8%) in our analysis.

Several limitations were present in this study. Lack of patient level data in place of hospitalizations is a major limitation; repeated hospitalizations for a patient cannot be identified and the clustering by patient is not taken into account. The potential confounding effect of individual income on the association between neighborhood affluence and type of procedure could not be assessed. There was also an absence of follow-up information to assess long-term outcomes and associated individual and societal costs.

To our knowledge, this is the first study to determine the effect of socioeconomic neighborhood profile on the utilization of laparoscopic over open colorectal procedures. While previous studies have examined pieces of the ecological model of health, this study has examined the combined effects of multiple spheres of influence in order to effectively allocate resources and suggest further study of the most influential domains. We look at the predictors, effect modifiers, and cost savings together for the first time in one study help to identify not only the magnitude of the differential in effectiveness, but also to quantify this difference according to hospital costs. These findings illustrate that differences in procedure utilization are also influenced by factors other than standard clinical and hospital characteristics. Larger cost savings could be a “side effect” driven by preferential minimally invasive procedure utilization for the wealthiest proportion of the population. Furthermore, with the burgeoning of personalized medicine, we must be sure to continue to recognize that distal epidemiologic determinants of health have an effect not only on patient health, but May also influence clinician behavior and financial outcomes. Our study brings awareness to the effect of socioeconomic status on clinical decision-making, and helps to address such disparities in order to improve the quality and equity of care among less affluent patients.

Conclusion

In conclusion, our study confirms that patients from high-income neighborhoods are more likely to undergo laparoscopic colorectal procedure. This preference was greater among men, younger adults, those who were free of cancer at hospitalization and those with little or no risk of mortality. Upstream social factors bias lower income patients towards a worse clinical presentation and lead to a lower cost open procedure because of the facility’s standard of care rather than just surgical risk factors. Our study highlights a troubling racial disparity and demonstrates the need for reducing socioeconomic-driven dissimilarities regarding patient care; this awareness should strengthen our resolve to improve the quality and equity of care among low-income patients.

References

Kim SJ, Choi BJ, Lee SC (2014) Overview of single-port laparoscopic surgery for colorectal cancers: past, present, and the future. World J Gastroenterol 20:997-1004.
Trinh BB, Nicole RJ, Adam TH, Tian H, Kandil E, et al. (2014) Robotic versus laparoscopic colorectal surgery. JSLS 18:4.
Yeo H, Niland J, Milne D, ter Veer A, Bekaii-Saab T, et al. (2015) Incidence of minimally invasive colorectal cancer surgery at National Comprehensive Cancer Network centers. J Natl Cancer Inst 107: 362.
Robinson CN, Chen GJ, Balentine CJ, Sansgiry S, Marshall CL, Anaya DA, Artinyan A, Albo D, Berger DH, et al. (2011) Minimally invasive surgery is underutilized for colon cancer. Ann Surg Oncol 18: 1412-1418
Wilson MZ, Hollenbeak CS, Stewart DB (2014) Laparoscopic colectomy is associated with a lower incidence of postoperative complications than open colectomy: a propensity score-matched cohort analysis. Colorectal Dis 16: 382-389.
Juo YY, Hyder O, Haider AH, Camp M, Lidor A, et al. (2014) Is minimally invasive colon resection better than traditional approaches?: First comprehensive national examination with propensity score matching. JAMA Surg 149 : 177-184.
Ohtani, H, Yutaka Tamamori, Yuichi Arimoto, Yukio Nishiguchi, Kiyoshi Maeda, et al. (2012) A meta-analysis of the short- and long-term results of randomized controlled trials that compared laparoscopy-assisted and open colectomy for colon cancer. J Cancer 3: 49-57.
Romano G, Gagliardi G, Bianco F, Parker MC, Corcione F. Corcione, et al. (2008) Laparoscopic colorectal surgery: why it is still not the gold standard and why it should be. Tech Coloproctol 12 :185-188.
Dik VK, Aarts MJ, Van Grevenstein WM, Koopman M, Van Oijen MG, et al. (2014) Association between socioeconomic status, surgical treatment and mortality in patients with colorectal cancer. Br J Surg 101: 1173-1182.
Scheppers E, van Dongen E, Dekker J, Geertzen J, Dekker J,et al. (2006) Potential barriers to the use of health services among ethnic minorities: a review. Fam Pract 23: 325-348.
Garg V, Raisch DW, Selig JP, Thompson TA (2013) Health disparities in clinical practice patterns for prostate cancer screening by geographic regions in the United States: a multilevel modeling analysis. Prostate Cancer Prostatic Dis 16: 193-203.
Bleday R, Babineau T, Forse RA (1993) Laparoscopic surgery for colon and rectal cancer. Semin Surg Oncol 9:59-64.
Ota DM, Nelson H, Weeks JC (1994) Controversies regarding laparoscopic colectomy for malignant diseases. Curr Opin Gen Surg 208-213.
Lacy AM, García-Valdecasas JC, Delgado S, Castells A, Taurá P, et al. (2002) Laparoscopy-assisted colectomy versus open colectomy for treatment of non-metastatic colon cancer: a randomised trial. Lancet 359: 2224-2229.
Guillou PJ, Quirke P, Thorpe H, Walker J, Jayne DG, et al. (2005) Short-term endpoints of conventional versus laparoscopic-assisted surgery in patients with colorectal cancer (MRC CLASICC trial): multicentre, randomised controlled trial. Lancet 365:1718-1726.
Buunen M, Veldkamp R, Hop WC, Kuhry E, Jeekel J, et al. (2009) Survival after laparoscopic surgery versus open surgery for colon cancer: long-term outcome of a randomised clinical trial. Lancet Oncol 10:44-52.
Guend H, Lee DY, Myers EA, Gandhi ND, Cekic V, et al. (2014) Technique of last resort: characteristics of patients undergoing open surgery in the laparoscopic era. Surg Endosc 29:2763-2769.
Zheng Z, Hanna N, Onukwugha E, Bikov KA, Mullins CD, et al. (2014) Hospital center effect for laparoscopic colectomy among elderly stage I-III colon cancer patients. Ann Surg 259:924-929.
Alnasser M, Schneider EB, Gearhart SL, Wick EC, Fang SH, et al. (2014) National disparities in laparoscopic colorectal procedures for colon cancer. Surg Endosc 28:49-57.
Robinson CN, Balentine CJ, Sansgiry S, Berger DH (2012) Disparities in the use of minimally invasive surgery for colorectal disease. J Gastrointest Surg 16:897-903.
Shy CM (1997) the failure of academic epidemiology: witness for the prosecution. Am J Epidemiol 145:479-484.
Poole C, Rothman KJ (1998) our conscientious objection to the epidemiology wars. J Epidemiol Community Health 52:613-614.
Sallis JF, ONFEB (2008) Ecological models of health behavior, in Health Behavior and Health Education: Theory, San Francisco.
Centers for Disease Control and Prevention (2014) The Social-Ecological Model: A Framework for Prevention.
Abenhaim HA, Azziz R, Hu J, Bartolucci A, Tulandi T (2008) Socioeconomic and Racial Predictors of Undergoing Laparoscopic Hysterectomy for Selected Benign Diseases: analysis of 341487 hysterectomies. Journal of Minimally Invasive Gynecology 15:11-15.
Iwashyna T, Curlin F, Christakis N (2002) Racial, ethnic, and affluence differences in elderly patients’ use of teaching hospitals. Journal of General Internal Medicine 17:696-703.
Chakravorty S (1996) a Measurement of Spatial Disparity: The Case of Income Inequality. Urban Studies 33:1671-1686.
HCUP Databases. Healthcare Cost and Utilization Project (HCUP).
Simorov A, Shaligram A, Shostrom V, Boilesen E, Thompson J, et al. (2012) Laparoscopic colon resection trends in utilization and rate of conversion to open procedure: a national database review of academic medical centers. Ann Surg 256:462-468.
Steele SR, Brown TA, Rush RM, Martin MJ (2008) Laparoscopic vs open colectomy for colon cancer: results from a large nationwide population-based analysis. J Gastrointest Surg 12:583-591.
Freeman PR (1993) the role of p-values in analysing trial results. Stat Med, 12:1443-1452.
Lemstra ME, Alsabbagh W, Rajakumar RJ, Rogers MR, Blackburn D (2013) Neighbourhood income and cardiac rehabilitation access as determinants of nonattendance and noncompletion. Can J Cardiol 29:1599-1603.
Hardman A, Ioannides YM (2004) Neighbors’ income distribution: economic segregation and mixing in US urban neighborhoods. Journal of Housing Economics 13:368-382.
Saez E (2009) striking it richer: The evolution of top incomes in the United States (update with 2007 estimates).
Dunlop DD, Manheim LM, Song J, Chang RW (2002) Gender and ethnic/racial disparities in health care utilization among older adults. J Gerontol B Psychol Sci Soc Sci 57:221-233.
Wood E, Sallar AM, Schechter MT, Hogg RS (1999) Social inequalities in male mortality amenable to medical intervention in British Columbia. Soc Sci Med 48:1751-1758.
Hamaker ME, Schiphorst AH, Verweij NM, Pronk A (2014) Improved survival for older patients undergoing surgery for colorectal cancer between 2008 and 2011. Int J Colorectal Dis 29:1231-1236.
Bianchi PP, Rosati R, Bona S, Rottoli M, Elmore U, et al. (2007) Laparoscopic surgery in rectal cancer: a prospective analysis of patient survival and outcomes. Dis Colon Rectum 50:2047-2053.
NCCN Clinical Practice Guidelines on Oncology (NCCN Guidelines), (2012) Rectal Cancer. Version 3.
Onega T, Duell EJ, Shi X, Wang D, Demidenko E, et al. (2008) Geographic access to cancer care in the U.S. Cancer. 112:909-918.
Fleshman J, Sargent DJ, Green E, Anvari M, Stryker SJ, et al. (2007) Laparoscopic colectomy for cancer is not inferior to open surgery based on 5-year data from the COST Study Group trial. Ann Surg 246:655-662.
Green BL, Marshall HC, Collinson F, Quirke P, Guillou P, et al. (2013) Long-term follow-up of the Medical Research Council CLASICC trial of conventional versus laparoscopically assisted resection in colorectal cancer. Br J Surg 100:75-82.
Sacerdote C, Ileana Baldi, Oscar Bertetto, Daniela DiCuonzo, Enzo Farina, et al. (2012) Hospital factors and patient characteristics in the treatment of colorectal cancer: a population based study. BMC Public Health, 12:775.
Bowman KC, Tabrizian P, Telem DA, Boudourakis L, Divino CM (2010) Health disparity in complicated colorectal cancer. Am Surg 76:164-167 .
Bianchi PP, Petz W, Luca F, Biffi R, Spinoglio G, et al. (2014) Laparoscopic and Robotic Total Mesorectal Excision in the Treatment of Rectal Cancer. Brief Review and Personal Remarks. Front Oncol 4:98.
Keller DS, Champagne BJ, Reynolds HL Jr, Stein SL, Delaney CP et al. (2014) Cost-effectiveness of Laparoscopy in Rectal Cancer. Dis Colon Rectum 57:564-9.

Follow Us

Share

Resources

Quick Links

Contact Us

Follow Us

Follow Us On: