Accuracy in 24-hour Urine Collection at a Tertiary Center

Original Research Accuracy in 24-hour Urine Collection at a Tertiary Center Carter Boyd, BS,1 Kyle Wood, MD,2 Dustin Whitaker, BS,1 Omotola Ashorobi, MD,2 Lisa Harvey, BS,2 Robert Oster, PhD,3 Ross P. Holmes, PhD,2 Dean G. Assimos, MD2 1University of Alabama-Birmingham School of Medicine, Birmingham, AL; 2Department of Urology, University of Alabama-Birmingham, Birmingham, AL; 3Department of Medicine, University of Alabama-Birmingham, Birmingham, AL There is a paucity of studies addressing the accuracy of 24-hour urine collection for assessing stone risk parameters. Collection accuracy is thought to be essential for assigning optimal therapy for stone prevention. The objective of this study was to determine factors associated with accurate and inaccurate collections. During a 2-year period (2015-2016), 241 stone formers completed 24-hour urine collections. They were divided into accurate collectors (AC), defined as at least one accurate urine collection, and inaccurate collectors (IC). Accuracy was assessed by 24-hour urine creatinine (Cr) excretion indexed to body weight (normal: males, 20-25 mg Cr/kg; females, 15-20 mg Cr/ kg). Demographic data analyzed included age, gender, race, insurance status, partner status, income, and education. Statistical analysis methods included the chi-square test, Fisher’s exact test, and the two-group t-test. Average age was 50.7 years at the time of collection; 50.2% were men, 86% were white, and 14% were black. Overall, 51.0% of collections were inaccurate. There was no statistical significance between AC and IC for gender (P 5 0.85), race (P 5 0.90), insurance status (P 5 0.85), recurrence (P 5 0.87), stone type (P 5 0.57), education (P 5 0.35), income (P 5 0.42), or poverty (P 5 0.35). Older age (P 5 0.017) and having a partner (P 5 0.022) were significantly associated with AC. The high rate of inaccurate 24-hour urine collections is a concern. The only factors we identified as influencing collection accuracy were age and partner status. These results underscore the importance of developing methods to improve the accuracy of collecting 24-hour urine samples. [Rev Urol. 2018;20(3):119–124 doi: 10.3909/riu0807] ® © 2018 MedReviews , LLC Vol. 20 No. 3 • 2018 • Reviews in Urology • 119 4170018_03_RIU0807_V3_ptg01.indd 119 10/29/18 6:01 PM Accuracy in 24-hour Urine Collection continued KEY WORDS Kidney stones • 24-hour urine collection, accuracy • Metabolic evaluation • Demographics • Socioeconomic factors T he prevalence of kidney stone disease has increased over the past several decades, now affecting 1 out of every 11 people in the United States.1 Stones can be prevented with the help of pharmacologic therapy and dietary modifications. Recurrent stone formers may benefit from metabolic evaluation by way of 24-hour urine stone risk parameter studies. This testing identifies specific metabolic abnormalities and enables the clinician to make individualized pharmacologic and dietary interventions to reduce the risk of future stone events.2 The role of 24-hour urine testing in urologic practice is important as both the American Urological Association (AUA) Guidelines and the European Association of Urology (EAU) Guidelines recommend 24-hour urine collections in high-risk stone formers. Additionally, the AUA recommends testing in interested low-risk stone formers.3,4 Although compliance with collections is important, so is the collection of accurate samples. Accuracy is typically assessed by 24-hour urinary creatinine excretion indexed to body weight. Accurate 24-hour urine collections decrease the need for additional urine collections and help the physician promptly identify any underlying abnormality and initiate targeted preventive therapy.2,4 There have been only two studies that we know of that assess accuracy of such collections and both demonstrated a high level of inaccuracy. In a retrospective study of 381 initial urine collections, Sawyer and colleagues reported 50.7% of urine collections to be inaccurate. In addition, 14% of patients had an overcollection and 37% of patients had an undercollection.5 In another retrospective study, McGuire and associates reviewed 1502 patients with 24-hour collections and found that 51% of patients submitted an inaccurate urine collection and that women were 50% more likely than men to submit an inaccurate urine collection.6 Although there has been some investigation of factors associated with patient compliance with 24-hour urine collections, this has not been done in great detail for accuracy.7 In this study, we aimed to identify factors that may influence 24-hour urine collection accuracy for stone-forming patients. Materials and Methods Institutional review board (IRB) approval was obtained. A retrospective chart review was performed of adult patients in our kidney stone clinic who completed 24-hour urine collection for urinary stone risk assessment in 2015 and 2016. The 24-hour urine collections are analyzed by an outside laboratory, Litholink®, which provides patients with instructions and materials for urine collection. The collected specimens are transported to the company by prepaid FedEx transport. Demographic information captured included age at collection, sex, race, insurance status (insured, non/underinsured), first-time stone former, recurrent stone former, stone composition (dominant stone component, >50%), partner status, and socioeconomic status. Insured patients included those with Medicare or third-party insurance coverage and non/underinsured included patients that lacked insurance (self-pay) or had Medicaid coverage. ZIP code of residence was used to estimate socioeconomic status and level of education. Data from the US Census Bureau’s most recent published demographic information on mean household income, percentage below the poverty line, and percentage high school education or above were utilized for these estimates.8,9 Partner status was assessed by chart review of social history. The factors were based on population modeling/ estimates. Thus, continuous variable analysis was not permitted. We classified their first 24-hour urine collection as either accurate or inaccurate based on 24-hour urinary Cr excretion indexed to body weight (normal: males, 20-25 mg Cr/kg; females, 15-20 mg Cr/kg) based on methodology used in prior publications.10 For patients with two collections, accuracy was defined as at least one accurate collection based on the 24-hour urinary Cr excretion indexed to body weight. Analyses were performed separately for the following two groups: overall accuracy and accuracy in individuals with two collections. Because less than 3% of the study population accounted for individuals that were neither white or black, these patients were excluded from the analysis. In addition, noninsured individuals and underinsured individuals were classified as non/underinsured due to small frequencies in the former two categories. Data obtained in 2015 and 2016 120 • Vol. 20 No. 3 • 2018 • Reviews in Urology 4170018_03_RIU0807_V3_ptg01.indd 120 10/29/18 2:57 PM Accuracy in 24-hour Urine Collection were combined into a single dataset. Descriptive statistics, including frequencies and proportions for categorical variables and means and standard deviations for continuous variables, were calculated for all study variables of interest. For categorical data, comparisons of proportions were performed using the chi-square test or Fisher’s exact test if the assumptions for the chisquare test were not tenable. For continuous data, comparisons of means were performed using the two-group t-test. It was determined using box plots, stem-and-leaf plots, normal probability plots, and the Kolmogorov-Smirnov test that the distributions of the continuous variables did not deviate greatly from a normal distribution. All statistical tests were twosided and were performed using a significance level of 5% (ie, a 5 0.05). Statistical analyses were performed using SAS software, version 9.4 (SAS Institute, Inc., Cary, NC). Results A total of 241 patients completed 24-hour urine collections in 2015 and 2016. The mean age of patients at collection was 50.7 years and the cohort had slightly more men (50.2%), than women (49.8%). Most of the patient population was white (white 86%, black 14%) and adequately insured (93.8%), whereas a smaller group was categorized as non/underinsured (6.2%). Recurrent stone formers comprised 80.1% of patients. More than half of all patients (51.9%) were calcium oxalate (CaOx) stone formers, whereas the remainder were either unknown or had other stone compositions. Most patients had domestic partners (partner 63.1%, no partner 36.9%). This patient population resides in ZIP codes that have an average high school graduation rate of 86.0%, a median household income of $51,569, and a poverty rate of 16.7%. Of the 241 patients that completed either one or two urine collections, the overall inaccurate collection rate was 51.0%, with accuracy defined as at least one accurate urine collection. Based on 24-hour urinary creatinine excretion indexed to body weight, 53.7% of inaccurate collections were undercollections whereas 46.3% were overcollections. Age (P 5 0.017) and partner status (P 5 0.022) significantly influenced collection accuracy (Table 1). The mean age was significantly different by nearly 5 years when comparing individuals with accurate collections (53.2 years) and inaccurate collections (48.3 years). The proportion of patients with accurate collections and a domestic partner (54.6%) was significantly greater than the proportion of patients with accurate collections without a partner (39.3%). TABLE 1 Overall Collection Accuracy Collection Accuracy Inaccurate Collection (n 5 123) Accurate Collection (n 5 118) P Value Age (years) 48.3 53.2 0.017 Sex    Male    Female 61 (49.6%) 62 (50.4%) 60 (50.9%) 58 (49.1%) 0.85 Race    White    Black 106 (86.2%) 17 (13.8%) 101 (85.6%) 17 (14.4%) 0.90 Partner status    No    Yes 54 (43.9%) 69 (56.1%) 35 (29.7%) 83 (70.3%) 0.022 Insurance status    Non/underinsured    Insured 8 (6.5%) 115 (93.5%) 7 (5.9%) 111 (94.1%) 0.85 Recurrent stone former    No    Yes 25 (20.3%) 98 (79.7%) 23 (19.5%) 95 (80.5%) 0.87 Stone type    CaOx    Non-CaOx 66 (53.7%) 57 (46.3%) 59 (50.0%) 59 (50.0%) 0.57 85.6% 86.5% 0.35 Income $50,484 $52,741 0.42 Poverty 17.3% 16.1% 0.35 Education Accuracy defined as at least one accurate collection. Vol. 20 No. 3 • 2018 • Reviews in Urology • 121 4170018_03_RIU0807_V3_ptg01.indd 121 10/29/18 2:57 PM Accuracy in 24-hour Urine Collection continued Regarding the remaining analyzed factors, there was no statistical significance between patients with accurate collections and inaccurate collections for sex (P 5 0.85), race (P 5 0.90), insurance status (P 5 0.85), recurrence (P 5 0.87), stone type (P 5 0.57), education (P 5 0.35), income (P 5 0.42), or poverty (P 5 0.35). One hundred and forty-nine patients collected two 24-hour specimens. When analyzing the ability of this group to collect at least one accurate specimen, this only occurred in 53.7% of patients. There was a significant difference (P 5 0.020) between the mean age of patients with accurate collections (54.0 years) and the mean age of patients with inaccurate collections (47.9 years) in those who collected two specimens. No other factors were impactful (Table 2). TABLE 2 Collection Accuracy for Individuals With Two Collections Collection Accuracy Inaccurate Collection (n569) Accurate Collection (n580) P Value Age (years) 47.9 54.0 0.020 Sex    Male    Female 32 (46.4%) 37 (53.6%) 40 (50.0%) 40 (50.0%) 0.66 Race    White    Black 59 (85.5%) 10 (14.5%) 66 (82.5%) 14 (17.5%) 0.62 Partner status    No    Yes 29 (42.0%) 40 (58.0%) 25 (31.3%) 55 (68.7%) 0.17 Insurance status    Non/underinsured    Insured 5 (7.3%) 64 (92.7%) 7 (8.7%) 73 (91.3%) 0.74 Recurrent stone former    No    Yes 23 (33.3%) 46 (66.7%) 19 (23.8%) 61 (76.2%) 0.19 Stone type    CaOx    Non-CaOx 38 (55.1%) 31 (44.9%) 34 (42.5%) 46 (57.5%) 0.13 85.3% 86.0% 0.61 Income $50,382 $51,852 0.68 Poverty 17.2% 16.8% 0.82 Education Accuracy defined as at least one accurate collection. Discussion Twenty-four-hour urine testing continues to be recommended for the identification of risk factors associated with kidney stones and to direct preventive therapy.3,4 Medical management of urolithiasis can diminish kidney stone recurrence, lessen the economic burden of such events, and potentially improve quality of life.11-16 Despite guideline directives to perform such testing, Milose and colleagues reported that 24-hour urine testing is done in only 7.4% of patients within 6 months of a kidney stone diagnosis.17 Kaplon and associates reviewed patients who presented to the emergency department with nephrolithiasis and found that only 23% of these patients received a metabolic workup.18 There is a paucity of studies evaluating what impacts accurate performance of these tests. Our combined inaccurate collection rate of 51.0% is like the generally cited rate in the literature of just more than 50% based on urinary Cr to weight ratios.5,6 Given that roughly half of inaccurate collections were undercollections and half were overcollections, it is possible that there was a lack of patient understanding on how to properly collect urine samples. These results may just be representative of a Gaussian distribution. However, other factors warrant consideration. The criteria em­ployed for assessing accuracy were based on urinary Cr indexed to body weight, which has certain limitations. These limitations include deviances associated with the effects of age, body size and composition, and renal function.5 What should be done with these inaccurate results? One may consider interpreting the study using a pro-rating strategy or repeating the test. 122 • Vol. 20 No. 3 • 2018 • Reviews in Urology 4170018_03_RIU0807_V3_ptg01.indd 122 10/29/18 2:57 PM Accuracy in 24-hour Urine Collection We found that older age and partner status significantly influenced collection accuracy. These results are similar to those of a previous study we conducted analyzing factors associated with patient compliance with 24-hour urine testing.7 According to the health belief model, older patients with a greater severity of illness are more motivated to comply with medical management.19 This may be relevant to kidney stone patients, as increased age is associated with increased number of potential stone events.20 Older patients with an extensive history of kidney stones may be more motivated to prevent stone recurrence and submit accurate urine samples. Another possible explanation is that older patients are more likely to have caregivers or partners at home to assist with medical regimens.19 Our results were somewhat different than those reported by Sawyer and colleagues, who found that younger individuals were prone to overcollection and older subjects to undercollection.5 We propose that patients with a domestic partner are more likely to receive accountability and encouragement to effectively and accurately complete 24-hour urine collections. In a review examining the relationship between social support and chronic illness selfmanagement, Gallant suggests that greater levels of social support are related to better self-management behaviors.21 In a meta-analytic review of 73 studies, DiMatteo found the odds of patient adherence to be 1.27 times higher in married patients when compared with unmarried patients and 1.38 times higher in those living with others when compared with those living alone.22 It is likely that having a domestic partner may increase patient adherence to medical treatment and healthcare management regimens. For individuals that collected twice, age remained significantly different between accurate and inaccurate groups. McGuire and associates found that women were more likely to submit an inaccurate collection.6 However, we did not identify this association and do not have an explanation for this finding. We did not assess for occupation in our study; something McGuire and colleagues found to impact accuracy.6 The Cr-to-body-weight assessment of 24-hour urine collection accuracy has recently been called into question. Leva and associates have proposed that the utilization of muscle mass may prove to be a better denominator. This group found that urinary Cr was more tightly correlated with muscle mass than body weight. Using the Cr-toweight ratios to access collection accuracy, 38% of 24-hour urine collections in their study were deemed inaccurate. The inaccuracy rate decreased to 15% for these same individuals when the Cr-to-leanbody-mass metric was employed.23 More studies are required to assess the effectiveness and utility of this parameter to measure accuracy of urine collections. Regardless of the metric used to assess accuracy, increased focus needs to be placed on instructing patients how to accurately collect. Methods of limiting patient inconvenience may be helpful.24 In addition, utilizing instructional videos in the clinic or online may prompt more accurate collections. The employment of financial incentives may also prove to be a useful motivational strategy.25-27 The results of inaccurate collections may still yield clinically relevant information. For example, if an individual who is classified as an overcollector has reduced volume and low citrate, therapy can be instituted. Similarly, if an undercollector has increased calcium, oxalate, or uric acid excretion, treatment directives can be made. Furthermore, urine pH and supersaturation indices would typically be valid in many cases. Limitations of this study include a relatively small sample size generated from the practice of a single surgeon. Furthermore, the bulk of patients cared for by this individual have complex stone problems and thus may not be reflective of patients seen in general urologic practice. Data were retrospectively obtained and were limited to patient information recorded in the electronic health record. The methods of defining educational and socioeconomic status also have some inherent limitations. In addition, Main Points • Collection accuracy is low and needs to be addressed and improved to increase the validity of 24-hour urine results. • The only factors identified as influencing collection accuracy were older age and partner status. • These results underscore the importance of developing methods to improve the accuracy of collecting such urine samples. Vol. 20 No. 3 • 2018 • Reviews in Urology • 123 4170018_03_RIU0807_V3_ptg01.indd 123 10/29/18 2:57 PM Accuracy in 24-hour Urine Collection continued health literacy could have played a role in how patients interpreted and followed collection instructions.28 Although we estimated that most subjects were high school graduates based on ZIP code of residence, this may not be indicative of a level of health literacy necessary to fully understand the provided written instructions. Another limitation is the definition used to define accuracy. Others have used different definitions of accuracy based on urinary Cr indexed to body weight.5,6 In addition, the urinary Cr indexed to body weight may in the future be supplanted by methods based on lean body mass. Conclusions Collection accuracy is low and needs to be addressed and improved to increase the validity of our 24-hour urine collection results. The only factors that we identified as influencing collection accuracy were older age and partner status. These results underscore the importance of developing methods to improve the accuracy of collecting such urine samples. References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. Scales CD Jr, Smith AC, Hanley JM, et al. Prevalence of kidney stones in the United States. Eur Urol. 2012;62:160-165. Hsi RS, Sanford T, Goldfarb DS, et al. The role of the 24-hour urine collection in the prevention of kidney stone recurrence. J Urol. 2017;197:1084-1089. Skolarikos A, Straub M, Knoll T, et al. Metabolic evaluation and recurrence prevention for urinary stone patients: EAU guidelines. Eur Urol. 2015;67:750. Pearle MS, Goldfarb DS, Assimos DG, et al. Medical management of kidney stones: AUA guideline. J Urol. 2014;192:316-324. Sawyer MD, Dietrich MS, Pickens RB, et al. Adequate or not? A comparison of 24-hour urine studies for renal stone prevention by creatinine to weight ratio. J Endourol. 2013;27:366-369. McGuire BB, Bhanji Y, Sharma V, et al. Predicting patients with inadequate 24- or 48-hour urine collections at time of metabolic stone evaluation. J Endourol. 2015;29:730-735. Boyd C, Wood K, Ashorobi O, et al. An intervention to increase 24-hour urine collection compliance. Urology Practice. 2018. doi: 10.1016/j.urpr.2018.01.002. Eisner BH, Sheth S, Dretler SP, et al. Effect of socioeconomic status on 24-hour urine composition in patients with nephrolithiasis. Urology. 2012;80:43-47. 2011-2015 American Community Survey 5-year estimates. United States Census Bureau website. https:// factfinder.census.gov/faces/nav/jsf/pages/index. xhtml. Accessed June 28, 2017. Ghiraldi EM, Reddy M, Li T, et al. Factors associated with compliance in submitting 24-hour urine collections in an underserved community. J Endourol. 2017;31(suppl 1):S64-S68. Sninsky BC, Nakada SY, Penniston DL. Does socioeconomic status, age, or gender influence appointment attendance and completion of 24-hour urine collections. Urology. 2015;85:568-573. Lotan Y. Economics and cost of care of stone disease. Adv Chronic Kidney Dis. 2009;16:5-10. Canvasser NE, Alken P, Lipkin M, et al. The economics of stone disease. World J Urol. 2017;35:1321-1329. Pearle MS, Calhoun EA, Curhan GC, et al. Urologic diseases in America project: urolithiasis. J Urol. 2005;173:848-857. Bensalah, K, Tuncel, A, Gupta, A, et al. Determinants of quality of life for patients with kidney stones. J Urol. 2008;179:2238-2243; discussion 2243. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. Penniston, KL, Antonelli, JA, Viprakasit, DP, et al. Validation and reliability of the Wisconsin Stone Quality of Life Questionnaire. J Urol. 2017;197:1280-1288. Milose JC, Kaufman SR, Hollenbeck BK, et al. Prevalence of 24-hour urine collection in high risk stone formers. J Urol. 2014;191:376-380. Kaplon DM, Sterrett S, Nakada SY. Medical management of acute urolithiasis in one American academic emergency room. BJU Int. 2010;105:856-858. Janz NK, Becker MH. The Health Belief Model: a decade later. Health Education Quarterly. 1984;11:1-47. Stamatelou KK, Francis ME, Jones CA, et al. Time trends in reported prevalence of kidney stones in the United States: 1976-1994. Kidney Int. 2003;63:18171823. Gallant MP. The influence of social support on chronic illness self-management: a review and directions for research. Health Educ Behav. 2003;30:170-195. DiMatteo MR. Social support and patient adherence to medical treatment: a meta-analysis. Health Psychol. 2004;23:207-218. Leva N, Sanford T, Hsi R, et al. Lean muscle mass is more accurate than creatinine to weight ratio to evaluate 24-hour urine collection adequacy: development and validation of a regression model. J Urol. 2017;197(supplement):e4-e5. Parks JH, Goldfisher E, Asplin JR, et al. A single 24hour urine collection is inadequate for the medical evaluation of nephrolithiasis. J Urol. 2002;167:16071612. Lange JN, Easter L, Amoroso R, et al. Internet program for facilitating dietary modifications limiting kidney stone risk. Canadian J Urol. 2013;20.5:69226926. Roberts CA, Geryk LL, Sage AJ, et al. Adolescent, caregiver, and friend preferences for integrating social support and communication into an asthma selfmanagement app. J Asthma. 2016;53:948-954. Noordraven EL, Wierdsma AI, Blanken P, et al. Financial incentives for improving adherence to maintenance treatment in patients with psychotic disorders (Money for Medication): a multicenter, open-label, randomized controlled trial. Lancet Psychiatry. 2017;4:199-207. Quinlan P, Price KO, Magid SK, et al. The relationship among health literacy, health knowledge, and adherence to treatment in patients with rheumatoid arthritis. HSS J. 2013;9:42-49. 124 • Vol. 20 No. 3 • 2018 • Reviews in Urology 4170018_03_RIU0807_V3_ptg01.indd 124 10/29/18 2:57 PM