Socioeconomic Disparities in the Acute Management of Stone Disease in the United States

Introduction

The prevalence of urolithiasis in the United States has nearly doubled over the past 20 years, from 5% of the population in 1994 to over 9% in 2012. Lower socioeconomic status has been significantly associated with increased stone episodes. ¹ This group of patients represents a demographic particularly at risk for health care disparities. Previously, Seklehner and colleagues assessed trends in the surgical management of ureteral calculi and found disparities based on age, gender, race, and geography. ² Similarly, Blackwell and colleagues demonstrated an increased delay in the treatment of infected urolithiasis in non-Caucasian patients. ³

The current standard of care for an acute stone episode in the setting of infection necessitates urgent renal decompression. However, for patients presenting with an episode of an acute, noninfectious stone, treatment modalities are dictated by various factors, including clinical setting and provider preference. While it is currently unknown whether socioeconomic and demographic factors influence treatment decisions in cases of noninfectious acute stones, we hypothesize that health care disparities exist in the treatment of these patients. In this study, we sought to determine the clinical and economic impact of treatment choice for the management of acute urolithiasis. We identified patients presenting to the hospital with an acute stone episode and assessed factors that may have influenced treatment decisions.

Materials and Methods

Patients and outcomes

International Classification of Diseases, Ninth Revision (ICD-9) codes were used to identify patients admitted from the ER with renal (592.0) or ureteral (592.1) calculi as the primary diagnosis. Treatment modality was assessed utilizing ICD-9 procedure codes to identify patients who underwent medical management, ureteral stent alone, ureteroscopy (URS), percutaneous nephrostomy (PCN), percutaneous nephrolithotomy (PCNL), or extracorporeal shockwave lithotripsy (SWL) (see Supplementary Table S1 for ICD-9 codes). Patients were divided based on type of intervention during their initial hospitalization, which included medical management (no stone procedure during admission), stent alone, URS, SWL, PCN, or PCNL.

Patient characteristics, including the presence of urinary tract infections (UTIs), sepsis, hydronephrosis, and acute renal insufficiency present on admission, were assessed. Additional patient variables included race, primary insurance provider, weekend admission, age, sex, income, and medical comorbidities. See Figure 1 for study design.

OPEN IN VIEWER

FIG. 1.

Study design. ER = emergency room; SWL = extracorporeal shockwave lithotripsy; PCN = percutaneous nephrostomy; PCNL = percutaneous nephrolithotomy; UTI = urinary tract infection.

Furthermore, a cohort of patients potentially eligible for upfront definitive stone removal was identified. This included patients admitted for renal or ureteral calculi in the absence of sepsis or UTI. Outcomes assessed for this cohort included postoperative sepsis, outpatient stone removal, 30-day readmissions, 90-day ER visits, and cost of the initial inpatient admission. Hospital costs were calculated by multiplying hospital charges by hospital-specific HCUP cost/charge ratios for the initial inpatient admission. Patients were excluded if they had a scheduled admission or were not admitted through the ER.

Results

We identified 146,199 patients who had an inpatient admission with primary diagnosis of kidney or ureteral stone in California (2007–2011) and Florida (2009–2014). The median age was 51.5 years (interquartile range 38–64). Patient race/ethnicity was Caucasian (61.6%), African American (6.5%), Hispanic (24.2%), Asian (2.2%), and other/unknown (5.5%). In terms of insurance status, 40.8% of patients were privately insured, 29.6% had Medicare, 16.9% had Medicaid, and 12.7% were self-pay/uninsured. Baseline patient characteristics are listed in Table 1. In addition to a diagnosis of kidney or ureteral stone at admission, 62.7% of patients had a concomitant hydronephrosis, 21% had UTI, 11.2% had acute renal insufficiency, and 1.9% were admitted with sepsis.

Overall, 45% of patients had no intervention at the time of their evaluation. Of the 55% of patients who underwent intervention, 42% received stent alone, 44% upfront URS, 1% PCN tube placement, 8% SWL, and 5% PCNL.

To determine which factors independently predict a patient's likelihood of undergoing upfront URS, we fit an adjusted multivariate model including variables significant on univariate analysis (Fig. 2). Patients <25 years (odds ratio [OR] = 0.82; 95% confidence interval [CI] = 0.77–0.88) were less likely to undergo upfront stone removal compared with those 25 to 50 years. In terms of racial differences, African Americans (OR = 0.92; 95% CI = 0.87–0.98) and Hispanics (OR = 0.9; 95% CI = 0.87–0.93) were significantly less likely to undergo upfront URS compared with Caucasians. Compared with patients having private insurance, those with Medicare (OR = 0.85; 95% CI = 0.82–0.88), Medicaid (OR = 0.79; 95% CI = 0.75–0.82), and the uninsured (OR = 0.82; 95% CI = 0.78–0.86) were significantly less likely to undergo upfront URS. Similarly, increasing median household income correlated with increased odds of undergoing upfront URS (upper quartile: OR = 1.2; 95% CI = 1.2–1.3). Patients admitted on the weekend were significantly less likely to undergo URS (OR = 0.95; 95% CI = 0.91–0.98). Patients with a Charlson comorbidity score ≥2 were less likely to undergo upfront URS (OR = 0.92; 95% CI = 0.88–0.96). Lastly, patients admitted to an academic center were more likely to undergo upfront URS (OR = 1.2; 95% CI = 1.1–1.2).

OPEN IN VIEWER

FIG. 2.

Multivariate regression model assessing impact of patient factors on the odds of undergoing upfront ureteroscopy. AKI = acute kidney injury; CCI = Charlson Comorbidity Index.

The clinical impact of upfront URS vs stent alone in the setting of a noninfectious acute stone episode was then investigated (Table 2). On multivariate analysis, we found that patients who underwent upfront URS had a lower likelihood of a 30-day readmission (OR = 0.7; 95% CI = 0.64–0.76), postoperative sepsis (OR = 0.63; 95% CI = 0.47–0.82), and, as expected, less likelihood of stone procedures within 90 days (OR = 0.1; 95% CI = 0.1–0.12). Cost analysis demonstrated that patients who underwent stent alone had significantly lower inpatient cost compared with upfront URS ($7476 vs $8673; p < 0.001).

Comments

Initial treatment options for the acute noninfected stone include upfront renal decompression, medical management, or definitive stone treatment. This study investigated the influence of patient-specific factors on treatment decisions for acute stones and the clinical impact of those decisions. Our study demonstrated that patients admitted with urolithiasis as the primary diagnosis were more likely to undergo upfront URS if they were wealthy, Caucasian, carried private insurance, presented on a weekday, or presented to an academic center. Furthermore, patients without a contraindication to upfront URS who underwent stent placement alone were more likely to have a readmission within 30 days, required a subsequent stone procedure within 90 days, or developed sepsis. The utilization of ureteral stent alone, foregoing upfront URS, represented a missed opportunity for intervention in noninfected acute stone episodes. Unfortunately, socioeconomic and temporal disparities exist, making early definitive intervention less likely for a subset of these patients.

Socioeconomic disparities in health care have been a well-documented phenomenon in medical literature. Lower socioeconomic status in the United States is associated with lower rates of preventative screening, ⁵ immunizations, ^6,7 later prenatal enrollment, poorer control of chronic diseases, ⁸ and higher rehospitalization rates for avoidable causes. ⁹ African Americans have been found to experience a particularly large disparity in care. They less frequently see specialists, receive less preventive care, undergo fewer costly technological procedures, and receive less intensive medical care overall. ^9,10 Many of these inequalities are seen in other minority groups as well, including Hispanics, Asians, and native Americans. Socioeconomic disparities have been demonstrated in urology patients as well, in regard to prostate cancer treatment. ^11,12 Friedlander and colleagues assessed variations in the use of definitive therapy among Caucasian and African American men with intermediate- and high-risk clinically localized prostate cancer using a nationwide database. They demonstrated that Caucasian men were significantly more likely to undergo definitive therapy than African Americans (83% vs 74%; p < 0.001). ¹² Additionally, a recent study reported the influence of race on timing for surgical decompression for the management of infected urolithiasis and found that African Americans and Hispanics were more likely to have delayed decompression (>48 hours after initial presentation) compared with Caucasians. ³ Our study expands the scope of racial/ethnic disparities in urology to the management of acute stones. African Americans and Hispanics were significantly less likely to undergo upfront definitive therapy. Additionally, we demonstrated a clear disparity with uninsured and publicly insured patients having a significantly lower likelihood of upfront URS compared with privately insured patients.

A contributing factor to the variability in management of patients with acute stones is lack of insurance and poor access to care. These issues are at the forefront of health care policy and are primary drivers contributing to racial and ethnic disparities. Efforts have been made to reduce the socioeconomic disparities in health care. Through the Affordable Care Act, access to care for patients with lower socioeconomic status has improved in both Medicaid and non-Medicaid expansion states. ¹³ Chen and colleagues demonstrated that after the implementation of the Affordable Care Act, the reduction in uninsured patients was most pronounced among African Americans and Hispanics. This has led to improved access to care, increased probability of visiting any physician, and a reduction in patients foregoing necessary care. ¹⁴ Further expansion of insurance coverage will help to reduce disparities in stone management by further improving access to care. While expanded insurance coverage is a first step, unconscious bias may also drive decision making. In our study, the uninsured patients were least likely to undergo upfront URS; however, the disparity also existed in the publicly insured. Improving access and promoting the realization of unconscious bias will increase our ability to deliver the same level of care to all patients.

In addition to socioeconomic disparities, we demonstrated a decreased likelihood of upfront stone removal if a patient was admitted on the weekend. The weekend effect has been documented in both general surgery and urology literature. Blackwell and colleagues assessed factors contributing to delayed intervention for infected acute urolithiasis. They demonstrated that weekend admissions influence time to intervention with a 26% reduction in a patient's odds of timely intervention compared with weekday admission. ³ Although the cause of weekend effect is unclear, it seems related to changes in practice patterns on the weekend. A possible explanation for this weekend effect in our study may be the lack of availability of necessary resources to perform definitive stone management. Taking this into consideration, individual hospitals must balance the costs of providing appropriate resources for urologists on weekends with the potential benefits of providing access to upfront stone treatment.

It is well established that patients with signs of infection should undergo timely renal decompression without attempted definitive stone treatment, given the risk of sepsis. However, in a well-selected patient, upfront treatment of urolithiasis is a safe and effective treatment strategy. The utility of emergent vs delayed treatment of urolithiasis has been assessed in the literature. Osorio and colleagues first assessed the utility of upfront URS in the treatment of noninfected, obstructing ureteral stones at a single institution. They demonstrated that upfront URS is safe and effective with 92.4% overall stone-free rate and overall complication rate of 4.2%. ¹⁵ Similarly, Picozzi and colleagues assessed the utility of upfront SWL as first-line treatment. They found stone-free rates of 79% and complication rates of 0.5% to 9.9%, which do not differ from international guidelines. ^16,17 Arcaniolo performed a meta-analysis of the use of SWL and URS as first-line treatment for ureteral stones at initial presentation and concluded that emergent SWL or URS is a safe and effective modality that should be offered to patients. ¹⁸ Our study similarly demonstrated URS to be safe with significant clinical benefits, such as avoidance of subsequent surgery and decreased 30-day readmissions.

Clark and colleagues ¹⁹ showed in 1993 that the annual cost of urolithiasis in the United States was estimated to be $1.83 billion. In 2012, estimates from the Urologic Diseases in America project suggested that aggregate expenditures for treating patients with kidney stones exceed $10 billion annually. ²⁰ The economic impact of urolithiasis includes cost of office and ER visits, hospital admissions, procedure, medications, and missed work days for the working-age population.

Given the large economic burden of stone treatment, it is important to critically evaluate our current practices to optimize cost-effectiveness. Our data demonstrate that upfront URS may have significant economic benefit, especially in the setting of a bundled payment model. We demonstrated that the cost of an admission with ureteral stent placement alone is $7476 vs $8673 for admission with upfront URS. Based on established billing data, a subsequent outpatient URS costs between $1739 and $3483, a clear area for cost saving. A successful upfront URS often avoids a second, costly operative procedure. While our data demonstrated a $1200 increase in cost for the initial admission when upfront URS is performed, clear savings exist by avoiding readmissions and secondary stone procedures. Unfortunately, the SASD and SEDD do not include charge or cost data, and we were therefore unable to assess the economics of subsequent procedures or ER visits within our dataset.

Several limitations exist given the use of a retrospective administrative database. The accuracy of the data relies on proper coding at each institution. We utilized previously established codes to improve accuracy and reproducibility of the data. Furthermore, as we excluded infected patients from our stent vs URS analysis, this relied on appropriate clinical documentation of UTI. In some cases, the surgeon may abort URS upon encountering purulence following ureteral catheterization. While this should be documented as UTI, many times this may not be appropriately captured. However, when assessing disparities in care, the incidence of a mischaracterized UTI should be uniform across patient variables. Additionally, access to appropriate equipment for definitive stone surgery must be taken into account. If a hospital does not have access to a laser or lithotripter in the urgent setting, definitive management is impossible. However, lack of access to appropriate equipment represents another disparity in access to care, and if there is a clinical and cost advantage to patients, effort must be made to provide access to appropriate equipment in all cases. Importantly, patient-level data are not included in the database. Therefore, analysis of variables such as white blood cell count, stone size and location, creatinine, and vital signs could not be assessed. In an effort to characterize patients, we utilized previously established ICD-9 codes for UTI, sepsis, and renal insufficiency. SWL was not included in our upfront stone treatment cohort given the relatively small number of patients and inability to report outcomes in this group due to data use restriction. Lastly, this dataset represents a subset of the US health care system. Therefore, the generalizability toward other systems and countries may be limited. Certain socioeconomic factors, namely insurance status, are not universally applicable. However, the racial and temporal data may be applicable more broadly.

Upfront URS is a safe and effective treatment option for patients admitted with urolithiasis in the absence of infection. Unfortunately, socioeconomic and temporal disparities exist in the utilization of this treatment modality, with minorities and poorly insured patients being less likely to undergo upfront definitive stone treatment. Furthermore, upfront treatment with URS has both clinical and economic advantages, compared with ureteral stent placement alone. Further investigation must assess which patients would be best suitable for upfront stone treatment. Prospective trials looking at various clinical factors, including stone location, renal function, and lab results, would be beneficial to determine which patients are best served by upfront treatment. Furthermore, a comparison of SWL vs URS for noninfected stones would be beneficial.

Conclusion

With evidence that socioeconomic disparities exist in the management of acute urolithiasis, urologists should offer appropriate treatment regardless of socioeconomic and insurance factors. Improved care for these at-risk populations may be achieved through insurance expansion and improved availability of the necessary resources to perform definitive stone treatment whenever possible. Upfront stone treatment in these clinical settings remains safe and effective; urologists must always consider the clinical and economic implications of deferring definitive stone treatment.