Insurance Status,Stone Composition,and 24-Hour Urine Composition

Abstract

Purpose:

We examined the stone composition, 24-hour urinary risk factors, and insurance status in patients evaluated in two regional stone clinics to further investigate the relationship between the socioeconomic status and kidney stone formation.

Materials and Methods:

We performed a retrospective review of stone formers who completed a 24-hour urinalysis as part of a metabolic evaluation for nephrolithiasis. Insurance status was determined by billing records and those with state-assisted insurance (SAI) were compared to patients with private insurance (PI). Multivariate analyses were performed adjusting for known risk factors for stones.

Results:

Three hundred forty-six patients were included. Patients with SAI (16%) were significantly more likely to be female (55% vs.38%, p=0.026) and younger (43.5 vs.49.2, p=0.003). Among those with stone composition data (n=200), SAI patients were as likely to form calcium phosphate (CaPhos) as calcium oxalate (CaOx) stones (46.9% vs.31.3%, p=0.44). PI patients were significantly less likely to form CaPhos than CaOx stones (10.1% vs.77.4%, p<0.001). On multivariate analysis, among those with calcium stones, the odds of forming CaPhos stones over CaOx stones were ten times higher among SAI patients compared to PI, odds ratio 10.2 (95% CI 3.6, 28.6, p<0.001). Further, patients with SAI had significantly higher urine sodium, pH, and supersaturation of CaPhos, and a lower supersaturation of uric acid compared to patients with PI.

Conclusions:

SAI was associated with a greater likelihood of a CaPhos stone composition and increased urinary risk factors for CaPhos stones. These findings may reflect dietary or other unmeasured differences, and have important implications for resource allocation and counseling, as treatment may differ for these groups.

Introduction

S ocioeconomic status (SES) is measured by using three main variables, including the income level, education level, and occupation. The epidemiologic association between the SES and disease prevalence has been studied across a broad spectrum of diseases. For example, there is a well-established inverse relationship between the risk of coronary heart disease and lower SES.^1,2 There is also support for this same relationship with chronic renal insufficiency, depression, diabetes, gastrointestinal disease, and the metabolic syndrome.^3,4 Adler and Ostrove further demonstrated that the socioeconomic health gradient applies to both morbidity and mortality.¹ This association has been referred to as the socioeconomic health gradient, which includes the traditional SES variables of income, education, and occupation as well as newer variables, such as increased psychological stressors, physical inactivity, smoking, and a poor dietary intake.³

Historically, the income level alone has been used to represent SES.¹ However, individual income levels can be difficult to ascertain and instead is often a gross estimation reflected as a percentage of a population within a certain area that lives below the poverty level. As Medicaid health insurance is a state-funded public health insurance program based on the financial need, it can be used as one way to estimate a patient's income level on an individual basis.

Approximately 13% of men and 7% of women in the United States will be diagnosed with a kidney stone at some point in their lives, and these numbers appear to be increasing.⁵ Specifically, the incidence of calcium phosphate (CaPhos) stones has increased over the last 30 years.⁶ Pearle published data on the important influences, such as age, sex, region, and race/ethnicity on the prevalence of urolithiasis within the United States.⁷ Another study by Saint-Elie et al suggested that stone formers of a lower SES had higher intake levels of all the daily dietary constituents with the exception of fat.⁸ More recent studies have reported diabetes and obesity as risk factors for stone formation.^9
–11 A recent study that examined the SES in a cohort of kidney stone formers noted that the income level (% of patients below the poverty line) and the education level (% of patients with less than high school education) for a given zip code were associated with significant increases in urine calcium.¹² Despite the evolving knowledge on risk factors for the development of kidney stones, little is known about the role that SES plays on the stone composition and a 24-hour urine composition.

In this study, we examine the relationship of the SES stone composition, and the 24-hour urine composition using insurance status as a surrogate for the SES within a cohort of known stone formers.

Materials and Methods

Study design

Internal Review Board approval was obtained before the study. A retrospective analysis was performed on a group of known stone-forming patients from two regional metabolic stone clinics (Lebanon, NH and Boston, MA). Patients who were greater than 18 years old and underwent a metabolic evaluation for stone disease were included in this study. The outpatient and inpatient hospital records were reviewed along with one 24-hour urinalysis from each patient. Patient demographic information included age, race, and body mass index (BMI) calculated as kg/m² from self-reported information at the time of the 24-hour urinalysis. A pertinent past medical history obtained through a chart review of the medical record included diabetes mellitus (DM), hypertension (HTN), and gout.

The stone type for each patient was determined based on the majority type (≥50%) as documented on the stone analysis done using infrared spectroscopy. Stone types represented in our cohort include calcium oxalate (CaOx), CaPhos, uric acid (UA), and cystine. When a stone was 50% CaPhos and 50% CaOx, it was considered a CaPhos stone, as has been done by other groups.¹³ CaPhos stones included calcium carbonate phosphate, hydroxyapatite, and brushite stones. No other stone types were encountered in this cohort. Health insurance records were obtained and each patient was categorized as having either state-assisted insurance (SAI) or private insurance (PI). PI was defined as any third party payers except federally sponsored insurance plans. Patients were excluded if they were younger than 18 years old, did not have complete medical records, or their 24-hour urinalysis were deemed inadequate (for men, 24-hour urine creatinine <800 mg; for women, 24-hour urine creatinine <600 mg). Patients with medicare (age based, not need based) and those without any insurance were excluded from the analysis.

Urine collection

All patients completed a 24-hour urine collection obtained in an outpatient setting under normal circumstances. A small aliquot from the urine sample was then sent to Litholink™ (Litholink Corp., Chicago, IL) for analysis. Metabolic analysis included calcium, citrate, sodium, creatinine, phosphorus, UA, oxalate, magnesium, potassium, sulfate, urea nitrogen, pH, protein catabolic rate (PCR) and urine volume. The iterative computer program EQUIL 2 was used to calculate the supersaturation ratios of CaPhos, CaOx, and UA.

Statistical analysis

In our analysis, we included only one 24-hour urinalysis per patient. If more than one 24-hour urinalysis was available, only the first one was used. The patients were split into two categories, SAI versus PI. Using two-sided t-tests with Fisher's exact correction when appropriate (for continuous variables) and chi-squared (for categorical variables), baseline characteristics were compared between the two groups, including age, gender, BMI, DM, HTN, and gout. Next, we used these same tests to examine univariate relationships between the insurance type and stone composition, as well as 24-hour urine composition. All variables with p<0.2 were entered into a multivariate model for predicting the stone type. Backward stepwise logistic regression was then used to generate a final predictive model. Model discrimination and calibration were tested using the area under the receiver operator curve (AUC) and goodness of fit (GOF) tests, respectively. All analyses were performed using STATA™12 software (Statacorp, College Station, TX).

Results

Baseline characteristics

Three-hundred forty-six patients were included. Patients with SAI comprised 16% of the cohort (n=55) and 84% (n=291) had PI. Patients with SAI were significantly more likely to be female (55% vs. 39%, p=0.026), and younger (43.5 vs. 49.2, p=0.003). There were no baseline differences between the two groups (SAI vs. PI) in mean BMI (29.8 vs. 28.5, p=0.18), or in the incidence of DM (7.3% vs. 7.2%, p=0.9), HTN (23.6 vs. 27.8%, p=0.52), or gout (9.1% vs. 4.8%, p=0.20, Table 1).

Table 1.

Patient Characteristics Among Stone Formers with Private Versus State-Assisted Insurance

Patient characteristic	PI (n=291)	SAI (n=55)	p-Value
Age, mean (SD)	49.2 (12.8)	43.5 (11.9)	0.003
Male, %	61.5	45.5	0.026
BMI, mean (SD)	28.5 (6.3)	29.8 (6.4)	0.176
Diabetes, %	7.2	7.3	0.988
Hypertension, %	27.8	23.6	0.521
Gout, %	4.8	9.1	0.201

SD=standard deviation; PI=private insurance; SAI=state-assisted insurance; BMI=body mass index.

Comparison of stone composition and insurance type

Overall, 200 patients (58%) had data pertaining to stone composition available for analysis. The majority of stones were CaOx (70%), followed by CaPhos (16%), UA (13%), and cystine (1%). Patients with SAI were as likely to form CaPhos as CaOx stones (46.9% vs. 31.3%, p=0.436). However, patients with PI were less likely to form CaPhos than CaOx stones (10.1% vs. 77.4%, p<0.001). When comparing stone types of patients with SAI versus PI, SAI patients had a significantly higher prevalence of CaPhos stones than PI patients (46.9% versus 10.1% p=0.02). We also found that patients with PI had a significantly higher prevalence of CaOx stones compared to patients with SAI (77.4% vs. 31.3% p=0.0013). There was no difference in the proportions of UA and cystine stones formed between the two groups (Fig. 1).

FIG. 1.

Proportion of different stone types formed by known stone formers with state-assisted insurance (left) and private insurance (right). CaPhos=calcium phosphate; CaOx=calcium oxalate.

Comparison of 24-hour urine composition between state-assisted and publically insured

Table 2 demonstrates the comparison of 24-hour urinalysis between patients with SAI and PI. We found significant differences in urine sodium (204.1 vs. 182.0, p=0.05), urine citrate (647.8 vs. 553.7, p=0.048), supersaturation of UA (0.7 vs. 1.0, p=0.015), supersaturation of CaPhos (1.7 vs. 1.3, p=0.005), and urine pH (6.3 vs. 6.0, p=0.002). Urine calcium was not significantly different (243.3 vs. 223.1, p=0.25). Similarly, there were no significant differences noted between urinary oxalate, volume, UA, potassium, magnesium, phosphate, ammonia, chloride, sulfate, urea nitrogen, PCR, or supersaturation of CaOx.

Table 2.

Comparison of 24-Hour Urinalysis Composition Between Patients with Private Versus State-Assisted Insurance

Urinalysis component	PI (n=291)	SAI (n=55)	p-Value
Sodium (mmol/day)	182.0	204.1	0.05
Citrate (mg/day)	553.7	647.8	0.048
Supersaturation of uric acid	1.0	0.7	0.015
Supersaturation of CaPhos	1.3	1.7	0.005
pH	6.0	6.3	0.002
Calcium (mg/day)	223.1	243.3	0.25
Oxalate (mg/day)	37.8	41.1	0.18
Volume (L/day)	2.0	2.0	0.56
UA (g/d)	0.7	0.7	0.14
Potassium (mmol/day)	66.0	72.4	0.13
Magnesium (mg/day)	102.3	107.9	0.37
Phosphate (g/day)	1.0	1.0	0.68
Ammonia (mmol/day)	38.7	37.8	0.73
Chloride (mmol/day)	175.3	192.5	0.12
Sulfate (mmol/day)	44.7	44.2	0.85
Urea Nitrogen (g/day)	12.3	12.2	0.91
PCR (g/kg/day)	1.1	1.1	0.64
Supersaturation of CaOx	6.9	6.9	0.98

CaPhos=calcium phosphate; CaOx=calcium oxalate; UA=uric acid; PCR=protein catabolic rate.

Predictors of CaPhos stone formation

On multivariate analysis, we examined independent predictors of CaPhos stone formation among only calcium-based stone formers (140 patients). Odds of forming CaPhos stones over CaOx stones are ten times higher for patients with SAI compared to those with PI (odds ratio [OR] 10.2, 95% CI 3.6, 28.6, p<0.001). Other independent predictors of CaPhos stone formation were BMI and female gender (Table 3). The final predictive model had excellent calibration (GOF p-value=0.2) and discrimination (AUC=0.8).

Table 3.

Multivariate Model for Predicting the Formation of Calcium Phosphate Stones Over Calcium Oxalate Stones

Covariate	OR	95% CI	p-Value
Male gender	0.2	0.1–0.6	0.002
BMI	1.1	1.1–1.2	0.031
SAI	10.2	3.6–28.6	<0.001

AUC: 0.80, GOF p-value=0.2.

OR=odds ratio; AUC=area under the receiver operator curve; GOF=goodness of fit.

Discussion

SES is a known risk factor for numerous acute and chronic diseases. Studies have shown a strong and consistent inverse relationship between SES and disease morbidity and mortality.^1,14 These studies show that a lower SES is associated with a higher incidence of diseases, such as cardiovascular disease, diabetes, metabolic syndrome, arthritis, tuberculosis, chronic respiratory disease, gastrointestinal disease, and adverse birth outcomes.^1,4 Other studies have shown the opposite relationship for diseases, such as breast cancer and melanoma, which demonstrate an increasing incidence of breast cancer and melanoma as the SES increases.^15,16 In all these studies, the SES was determined by income, education, and occupation individually or in combination. All of these surrogates for SES have been used interchangeably and are noted to be closely correlated with one another.^14,17 Combined, these studies demonstrate that both a higher and lower SES can be risk factors for developing certain diseases.

Historically, calcium stones were considered a disease of the affluent and attributed to the increased food-purchasing capacity.¹⁸ Moreover, we evaluated SES and its association with the risk of kidney stone formation by comparing both the education and poverty level to a 24-hour urinalysis and known risk factors for stone formation. This study showed that the decreasing local education level and the increasing local poverty levels resulted in increased levels of urine calcium excretion. The study concluded that lower levels of education and increasing poverty independently lead to increased urinary excretion of calcium.¹⁹

In our cohort of stone formers, we used the health insurance type as a reflection of the individual patient's income level. SAI is issued on a financial need basis and thus serves as a surrogate for a lower SES. Conversely, PI may serve as a surrogate for a higher SES. We compared each patient's stone type within the two SES groups and evaluated the 24-hour urinalysis and known metabolic risk factors for the formation of each stone type within these two groups.

The results of our study show a significant difference in the prevalence of stone types between the two groups. Patients with SAI formed CaPhos and CaOx stones in equal proportions, whereas those with PI formed CaOx stones much more commonly than CaPhos. Further, we show that SAI alone was a risk factor for developing CaPhos stones with an OR of 10.2. To our knowledge, this is the first study to show a significant relationship between the SES and stone composition. The prevalence of UA, cysteine, and magnesium phosphate stones were too low in this cohort to calculate a difference.

Further, the stone prevalence we report correlates with results of our multivariate analysis of the 24-hour urines. We show that SAI is associated with higher levels of urine sodium, supersaturation of CaPhos, and a higher pH. These factors have been implicated as known risk factors for CaPhos stone formation.^{12,20

–24} Additionally, we found that PI is associated with higher levels of urine supersaturation of UA, a lower urine pH, and lower citrate levels, which are also known risk factors for the development of CaOx stones.^13,25,26 These results support the data we report on stone prevalence among the two socioeconomic groups and provide a physiologic basis for the observed clinical findings.

We also found in our study that SAI is associated with higher concentrations of urine sodium, supersaturation of CaPhos, urine citrate, and urine pH and a decrease in supersaturation of UA. One may suspect that patients from the SAI group consume less animal protein and higher amounts of dietary sodium. Several studies have shown that high levels of sodium excretion lead to higher urinary calcium levels.^24,27 Sakhaee et al further showed that increased dietary sodium not only increases calcium excretion in the urine, but also increases the urine pH and supersaturation of CaPhos.²⁸ More recently, it has been suggested that increased urinary sodium actually decreases the patient's risks of CaOx stone formation by decreasing the urinary supersaturation of CaOx concentrations.²⁰ Finally, Parks et al reported that both increasing urinary pH and citrate levels are associated with an increasing risk of CaPhos stone formation, which is reflected in our data.²²

Interestingly, in our cohort, along with a higher prevalence of CaPhos stones, the SAI group was also younger and more frequently female. The etiology behind these demographic differences are unknown, however, these findings are consistent with the findings of a study by Parks et al who reported that among CaPhos stone formers, the female gender predominated. Similar to our study, they also found that CaPhos stone disease began at a younger age.²³ While the correlation between the two studies is compelling, more studies are needed to elucidate the mechanism behind these relationships.

One trend that is becoming clear is the progressive increase in the number of CaPhos stones over the last 30 years.^6,23 There is evidence to suggest that patients with CaPhos stones require more procedures than those with CaOx stones to be rendered stone free.^23,29,30 Thus, our findings of increased CaPhos stone prevalence in those with SAI may have broader implications for resource allocation and stone prevention strategies for this group. Whether this increased observation is related to changes in diets over the last 30 years, increase in the socioeconomic disparity, increasing access for patients with SAI to stone clinics, or some other variable is as of yet unknown. Further studies exploring the impact of SES on nephrolithiasis are necessary.

We recognize several limitations inherent in a retrospective study. Within our study population, we may be seeing a presentation bias of patients with SAI, and thus, only capturing compliant patients with adequate follow-up to at least complete a 24-hour urinalysis. The disproportionate representation between the two groups may not reflect the true differences between the two populations. We also did not have the stone type for every patient who was included in our study, which may have skewed our stone composition analysis between the two groups. Further, intrinsic metabolic factors cannot be ascertained from this study as the diet has not been controlled. However, due to this, we were able to uncover factors related to self-selected diets, which likely differ between the two groups.

Conclusion

In stone formers, the proportion of CaPhos stones relative to CaOx stones is significantly higher in those with SAI than those with PI. These findings were substantiated in our analysis of the two groups' 24-hour urine collections, which demonstrated abnormalities associated with an increased risk for their associated stone type. While the underlying etiology of these differences remains unknown, one factor may be the intrinsic differences in diets between the two groups. These findings may have implications in terms of the cost of treating patients with SAI and the allocation of counseling services and resources, as the best approach to counseling patients for stone prevention may differ between populations.

Footnotes

Acknowledgments

Data collection was, in part, done by Sonali Sheth and Corey Armstrong. Further, this essay was the recipient of the 1st place in the 2012 Max K. Willscher Resident Research Award Competition, an essay competition among urology residents within the New England section of the AUA.

Disclosure Statement

No competing financial interests exist.

Abbreviations Used

References

Adler

, Ostrove

. Socioeconomic status and health: What we know and what we don't. Ann N Y Acad Sci, 1999; 896:3–15.

Fiscella

, Tancredi

, Franks

. Adding socioeconomic status to Framingham scoring to reduce disparities in coronary risk assessment. Am Heart J, 2009; 157:988–994.

James

, Nelson

, Ralph

, Leather

. Socioeconomic determinants of health. the contribution of nutrition to inequalities in health. BMJ, 1997; 314:154–159.

Tamashiro

. Metabolic syndrome: Links to social stress and socioeconomic status. Ann NY Acad Sci, 2011; 1231:46–55.

Stamatelou

, Francis

, Jones

, Nyberg

, Curhan

. Time trends in reported prevalence of kidney stones in the United States: 1976–1994. Kidney Int, 2003; 63:1817–1823.

Mandel

, Mandel

, Fryjoff

, Rejniak

, Mandel

. Conversion of calcium oxalate to calcium phosphate with recurrent stone episodes. J Urol, 2003; 169:2026–2029.

Pearle

. Urologic diseases in America project: Urolithiasis. J Urol, 2005; 173:848–857.

Saint-Elie

, Patel

, Healy

, Solomon

, Pattaras

, Qian

et al. The impact of income and education on dietary habits in stone formers. Urology, 2010; 76:307–313.

Eisner

, Porten

, Bechis

, Stoller

. Diabetic kidney stone formers excrete more oxalate and have lower urine pH than nondiabetic stone formers. J Urol, 2010; 183:2244–2248.

10.

Semins

, Shore

, Makary

, Magnuson

, Johns

, Matlaga

. The association of increasing body mass index and kidney stone disease. J Urol, 2010; 183:571–575.

11.

Taylor

, Stampfer

, Curhan

. Obesity, weight gain, and the risk of kidney stones. JAMA, 2005; 293:455–465.

12.

Asplin

, Parks

, Lingeman

, Kahnoski

, Mardis

, Lacey

et al. Supersaturation and stone composition in a network of dispersed treatment sites. J Urol, 1998; 159:1821–1825.

13.

Breslau

, Brinkley

, Hill

, Pak

. Relationship of animal protein-rich diet to kidney stone formation and calcium metabolism. J Clin Endocrinol Metab, 1988; 66:140–146.

14.

Ostrove

, Adler

. The relationship of socioeconomic status, labor force participation, and health among men and women. J Health Psychol, 1998; 3:451–463.

15.

Devesse

, Diamond

. Association of breast cancer and cervical cancer incidences with income and education among whites and blacks. J Natl Cancer Inst, 1980; 65:515–528.

16.

Hakeama

, Hakulinen

, Pukkala

, Saxen

, Teppo

. Risk indicators of breast cancer and cervical cancer on ecologic and individual levels. Am J Epidemiol, 1982; 116:990–1000.

17.

Winkleby

, Hatulis

, Frank

, Fortmann

. Socioeconomic status and health: How education, income, and occupation contribute to risk factors for cardiovascular disease. Am J Public Health, 1992; 82:816–820.

18.

Robertson

, Peacock

, Heyburn

, Hanes

. Epidemiological risk factors in calcium stone disease. Scand J Urol Nephrol, 1980; 53:15–30.

19.

Eisner

, Sheth

, Dretler

, Herrick

, Lang

, Pais

Jr.

Impact of socioeconomic status on 24-hour urine composition in patients with nephrolithiasis. Urology, 2012; 80:43–47.

20.

Eisner

, Eisenberg

, Stoller

. Impact of urine sodium on urine risk factors for calcium oxalate nephrolithiasis. J Urol, 2009; 182:2330–2333.

21.

Pak

, Waters

, Arnold

, Holt

, Cox

, Barilla

. Mechanism for calcium urolithiasis among patients with hyperuricosuria: Supersaturation of urine with respect to monosodium urate. Clin Invest, 1977; 59:426–431.

22.

Parks

, Coward

, Coe

. Correspondence between stone composition and urine supersaturation in nephrolithiasis. Kidney Int, 1997; 51:894.

23.

Parks

, Worchester

, Coe

, Evan

, Lingeman

. Clinical implications of abundant calcium phosphate in routinely analyzed kidney stones. Kidney Int, 2004; 66:777.

24.

Breslau

, McQuire

, Xerwekh

, Pak

. The role of dietary sodium on renal excretion and intestinal absorption of calcium and vitamin D metabolism. J Clin Endocrinol Metab, 1982; 55:369.

25.

Coe

, Kavalach

. Hypercalciuria and hyperuricosuria in patients with calcium nephrolithiasis. N Engl J Med, 1974; 281:1344–1350.

26.

Pais

, Holmes

, Assimos

. Effect of dietary control of urinary uric acid excretion in calcium oxalate stone formers and non-stone-forming controls. J Endourol, 2007; 21:232–235.

27.

Kok

, Iestra

, Ddoorenbos

, Papapoulos

. The effects of dietary excesses in animal protein and in sodium on the composition and the crystallization kinetics of calcium oxalate monohydrate in urines of healthy men. J Clin Endocrinol Metab, 1990; 71:861–867.

28.

Sakhaee

, Harvey

, Padalino

, Whitson

, Pak

. The potential role of salt abuse on the risk for kidney stone formation. J Urol, 1993; 150:310–312.

29.

Kacker

, Meeks

, Zhao

, Nadler

. Decreased stone-free rates after percutaneous nephrolithotomy for high calcium phosphate composition kidney stones. J Urol, 2008; 180:958–960.

30.

Krambeck

, Handa

, Evan

, Lingeman

. Profile of the brushite stone former. J Urol, 2010; 184:1367–1371.