Printer Friendly

A comparison of reagent strips and the refractometer for measurement of urine specific gravity in hospitalized children.

Pediatric nurses use specific gravity to assess hydration status in children. Clinical decision-making may be affected by specific gravity. For example, a specific gravity above 1.020 may result in chemotherapy being delayed until the child is better hydrated, a nurse increasing an IV rate, or a child being encouraged to drink more fluids.

Specific gravity is most commonly measured by one of three methods: hydrometry, refractometry, or reagent strip. Urine hydrometry is a method to test urine specific gravity by using a hydrometer suspended in a graduated vessel. The vessel is filled with 20 ml of urine and the hydrometer is floated in the urine. The hydrometer has a series of numbers representing the value for the specific gravity. The number on the hydrometer at the meniscus of the liquid represents the urine specific gravity. Hydrometry is not practical in young children because an adequate amount of urine cannot be collected, especially in diapered children (Lloyd, 1993). A refractometer measures urine specific gravity by analyzing the amount of light passing through a drop of urine on a glass plate. The tiny amount of urine required for the test makes refractometers very useful in pediatrics. Reagent strips consist of a plastic strip with small absorbent reagent squares. Reagent strips measure urine glucose, blood, protein, ketones, pH, bilirubin, urobilinogen, nitrite, ieukocytes, and specific gravity. They require a small amount of urine and can be pressed into urine voided in a diaper.

In one institution, standard practice was to measure specific gravity using the refractometer. Reagent strips were used to measure urine pH and all other substances. The purpose of this study was to determine if the specific gravity result obtained by reagent strip was equivalent to the result obtained by refractometer. An additional purpose of this research was to determine if an adequate sample of urine for reagent strip testing could be obtained from a wide age range of pediatric clients.

Review of Literature

A review of literature on bedside testing for urine specific gravity was conducted using the following databases: CINAHL, Grateful Med, and Medline. Gounden and Newall (1983) conducted a study using urine from 12 adults to test the relationship between urine specific gravity measured by hydrometry, reagent strip, and refractometer. They found a correlation of .91 between refractometer and reagent strip results. Another study of urine specific gravity obtained from adult samples compared various diaper brands and collection methods (Gammage & Yarandi, 1993). They reported that it was difficult to aspirate urine from a diaper containing less than 40 mi. Specific gravity results were found to be affected by diaper brand, urine volume in the diaper, time interval between void and urine testing, and method used to measure specific gravity. Specific gravity measured by refractometer immediately after a void was found to be the same whether in a container or diaper. The authors continued to measure the specific gravity over time and found that the specific gravity rose as the urine remained in the diaper longer. The specific gravity obtained by reagent strip was significantly higher when compared to container urine when the urine remained in the diaper longer than 4 hours. A review of practice and research by Zaloga (1993) concluded that the refractometer was more accurate than the reagent strip in measuring urine specific gravity in adults. Guthrie, Lott, Kriesel, and Miller (1987) tested urine samples from 279 adults. Urine specific gravity was measured using pycnometry as the "gold standard." The results from pycnometry were compared with those obtained by refractometry and reagent strip. A correlation of .88 was reported between results obtained by refractometry and reagent strip. The authors reported that the reagent strip values were generally lower than those obtained by refractometry.

The review of the literature suggested that refractometer and reagent strip values were closely related. However, the literature did not provide clear direction for adopting either method in clinical practice. There was also a dearth of studies of urine obtained from children making it difficult to choose one method over another for measuring urine specific gravity in this population.


A convenience sample of at least 100 was necessary based on a power analysis with a small effect size for a p [is less than] .005 (Cohen, 1987). The convenience sample was drawn from clients requiring measurement of specific gravity on two inpatient pediatric units: neonatal intensive care and hematology-oncology-general pediatrics.

Each urine sample was tested twice, once using the refractometer and once using the reagent strip. For the reagent strip, manufacturer's instructions were followed. Specific gravity was read at 45 seconds (Miles Laboratory, 1994). The refractometer measures the index of refraction of the test sample, with zero representing the value for distilled water. The manufacturer states there are no critical adjustments of the instrument for measuring urine specific gravity (Kernco Instruments, 1995). All nurses were required to demonstrate competency in urine testing during orientation.

For the infants, reagent strips were placed directly on the wet diaper and urine was aspirated from the diaper for refractometer readings. Nurses were requested to test each sample twice, once using each method. Nurse participation was voluntary. The nurse recorded the reading from each method, the ability to obtain an adequate sample, and any other problems collecting or testing the sample. A cover letter explaining the study and a data collection tool were posted in the utility room on both units. Exempt status was obtained from the Institutional Review Board. Parental consent was not necessary because there were no patient identifiers associated with the data and testing was performed on urine collected during routine patient care.


A sample of 157 specimens was obtained over a 5-week study period. The age range of children from whom urine was tested was 1 day to 16 years. Samples were obtained via U-bag, urine catheterization, free voiding, and diaper. Nurses reported few problems obtaining a sample. Nurses reported trouble distinguishing between two adjacent colors on the reagent strip for three samples. These three samples were not included in data analysis. Three other readings included the comment: "45 seconds is too long," noting that the immediate reading of the refractometer was preferred.

A Bland-Altman plot (see Figure 1) was used to determine the extent of agreement for urine specific gravity measurement between the refractometer and the reagent strip. Bland-Altman was chosen over correlation and linear regression methods because the data were expected be highly correlated (Bland & Altman, 1986).


The Bland-Altman plot for the data shows a high degree agreement between the two methods of analysis. The plot also shows that the methods agree for a wide range of values of urine specific gravity. Although most specific gravities obtained fell from 1.005-1.020, the plot shows there was agreement even outside that range.

A linear regression scatterplot of the data revealed not only that the reagent strip reading was highly correlated with the refractometer but that the reagent strip read significantly higher than the refractometer (p [is less than] .0001).


The study found that comparable results were obtained in pediatric urine tested by refractometer and by reagent strip. The Bland-Airman plot was useful for evaluating the agreement between the two measurement methods for urine specific gravity. Correlation and linear regression are prevalent in method-comparison studies but are not always the best for determining how well two methods of clinical measurement compare (Szafiarski & Slaughter, 1996). Bland and Altman (1986) point out that in clinical practice it is very unlikely that different methods will yield identical results. Decisions about adopting a new method of measurement in clinical practice depend on the how much the new method is likely to differ from the old. Bland and Altman (1986) argue that if the difference is not enough to cause a problem in clinical decision-making, then a new method may replace an old method. In addition to the findings of agreement between the two methods, nurses reported few problems obtaining samples from children of all ages regardless of the testing method. In cases where a child needs only a urine specific gravity measurement, the refractometer is faster and less costly.

Cost savings for choosing one method over another may be site specific. The cost of multiple additional refractometers could be as much as $1200 per year, particularly in one setting where an addition of 10,000 square feet, including 20 additional patient rooms, was recently completed. The cost of a bottle of reagent strips is approximately $35. One facility's policy is to charge a bottle of reagent strips to all patients requiring urine tests in addition to specific gravity.

The focus of this study was to determine the agreement between the two specific gravity assessment methods and not to determine which method was most accurate. Information about the source of the urine sample, whether diaper, free void, or urinary catheter was not obtained. The time between patient void and urine testing was not recorded. Future studies considering these variables may be warranted as Gammage and Yarandi (1993) reported specific gravity measurements by refractometer were inaccurate within 1 hour after voiding and inaccurate 4 hours after voiding when measured by reagent strips. The manufacturer of reagent strips recommended urine be tested immediately after voiding (Miles Laboratory, 1994).


Because the linear regression model showed the reagent strip readings to be higher, staff felt that by using the reagent strip they might be making a decision that ensured adequate hydration for patients receiving chemotherapy. Since the readings from the reagent strip were higher, the nurse would be making a more conservative decision in the determination of patient hydration status.


Bland, J., & Altman, D. (1986). Statistical methods for assessing agreement between two methods of clinical measurement. The Lancet, 307-310.

Cohen, J. (1987). Statistical power analysis for the behavioral sciences. Hillsdale, NJ: Lawrence Erlbaum Associates.

Gammage, D., & Yarandi, H. (1993). The effects of diaper brands, urine volume, and time on specific gravity measurement. Journal of Pediatric Nursing, 8, 10-14.

Gounden, D., & Newall, R. (1983). Urine specific gravity measurements: Comparison of a new reagent strip method with existing methodologies as applied to the water concentration/dilution tests. Current Medical Research and Opinion, 8, 375-381.

Guthrie, R., Lott, J., Kriesel, S., & Miller, I. (1987). Does the dipstick meet medical needs for urine specific gravity? The Journal of Family Practice, 25, 512-514.

Kernco Instruments. (1995). Clinical refractometers. El Paso, TX: Kernco, Inc.

Lloyd, C. (1993). Making sense of reagent strip urine testing. Nursing Times, 89(48), 34-36.

Miles Laboratory. (1994). Package insert for Multistix[R] SG reagent strip. Elkhart, IN: Miles, Inc.

Szaflarski, N., & Slaughter, R. (1996). Technology assessment in critical care: Understanding statistical analyses used to assess agreement between methods of clinical measurement. American Journal of Critical Care, 5, 207-216.

Zaloga, G. (1993). Reagent testing: Rapid, accurate urine testing at the bedside. Consultant, 33(6), 90-92, 95-98.

The Practice Applications of Research section presents reports of research that are clinically focused and discuss the nursing application of the findings. If you are interested in author guidelines and/or assistance, corntact Janice S. Hayes, PhD, RN; Section Editor; Pediatric Nursing; East Holly Avenue Box 56; Pitman, NJ 08071-0056; (609) 256-2300 or FAX (609) 256-2345.

Sharon Jackson Barton, PhD, RN, is Assistant Professor and Nurse Researcher, University of Kentucky and UK Children's Hospital, Lexington, KY.

Sharon Sallee Holmes, MSN, RN, is Division Director, University of Kentucky Children's Hospital, Lexington, KY.
COPYRIGHT 1998 Jannetti Publications, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1998 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Barton, Sharon Jackson; Holmes, Sharon Sallee
Publication:Pediatric Nursing
Date:Sep 1, 1998
Previous Article:Fighting Chance.
Next Article:Effects of television viewing on children's development.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters