Determination of the solids retainment effectiveness of disposable swim diapers.

Introduction

The contamination of public swimming and bathing facilities by fecal material associated with infants and toddlers in diapers is becoming a significant public health concern in the United States. Public health officials and pool facility managers are increasingly faced with the conflict between providing accessible aquatic recreation opportunities for infants and young children and protecting the health of all pool users. This dilemma has become particularly acute at urban public facilities serving primarily low-income families where the exclusion of infants and toddlers effectively curtails the ability of a parent to bring the family to the pool.

In June of 1998, 27 children in the Atlanta metropolitan area were infected with pathogenic E. coli bacteria at a local water slide park in Cobb County, Georgia (Gilbert & Blake, 1998). The contamination that led to these illnesses was traced to the kiddie pool at the White Water facilities; it resulted in seven hospitalizations and the death of a two-year-old girl. It is believed that the original source of the E. coli may have been tainted ground beef produced in April of 1998.

Outbreaks of recreational water-borne disease appear to be increasing. For example, the Centers for Disease Control and Prevention (CDC) found that 36 recreational water-borne disease outbreaks involving gastroenteritis occurred during 1999-2000, twice the number that occurred during 1997-1998 (Lee, Levy, Craun, Beach, & Calderon, 2002). The most common types of outbreaks were Cryptosporidium parvum in swimming pools or interactive fountains and Escherichi coli O157:H7 in freshwater venues. Because diarrheal illness is underreported to public health authorities, the number of reported outbreaks associated with recreational water use is probably underestimated by these reports (Mead et al., 1999). The recent increase in reported outbreaks is probably due to improved surveillance and reporting as well as to a true increase in the number of such outbreaks (Lee et al., 2002). More recently, a large outbreak of shigellosis involving 45 primary cases occurred at an unchlorinated fill-and-drain wading pool in Iowa (Lohff et al., 2001).

Some state health departments have implemented policies to address this increasing public health risk. For example, the State of Illinois Swimming Pool and Bathing Beach Code now requires "tight fitting rubber or plastic pants to be worn by infants at pools" (Metz, 2000). Also, CDC has set up a Web site informing the public about waterbornedisease risks from recreational swimming (CDC, 2003).

The many different types of bacteria commonly associated with fecal-coliform measurements breed and thrive within the digestive system of warm-blooded animals. Although most of these bacteria are harmless and many actually aid in digestion and other metabolic processes, several types of fecal bacteria can have extremely detrimental health effects if they are consumed. In many instances, ingestion of contaminated water results in lengthy gastrointestinal illness. In the most extreme cases, or without proper medical attention, ingestion of these pathogens can be fatal.

Fecal contamination of drinking-water sources or recreational areas can result in infection by Salmonella, Giardia, Cryptosporidium, and, as in the Atlanta case, forms of mutagenic E. coli. It is important to remember that in some cases the ingestion of one single protozoan or bacterium can be enough to infect an individual. For these reasons the national standard for fecal coliform in drinking water is less than one colony per 100 milliliters (mL) and 200.0 colonies per 100 mL for whole-body-contact recreation uses. Infants and toddlers (as well as the elderly and people with compromised immune systems) are especially vulnerable to these types of health threats.

One possible solution to this widespread public health concern is the development and required use of diaper-type swimwear, which might effectively prevent the release of fecal material during normal aquatic play activity. Thus, the purpose of this research project was to begin to investigate the potential for specially designed disposable swim diapers, also called swim pants, to prevent fecal contamination of swimming pools and other whole-body-contact recreation areas. The experiments were designed to determine the ability of the swim pants to retain fine particulate matter under simulated water play conditions. Attempts were made to simulate 1) the best possible fit of the swim pants on a young child and 2) the active play conditions a toddler would encounter in real-life "kiddie pool" and swimming pool situations.

Methods

Three commercially available brands of swim pants were tested in these experiments: Luvs "Splashwear" by Proctor and Gamble, Huggies "Little Swimmers" by Kimberly-Clark, and "Swimdiapers" by IPlay, Inc. The sizes selected for testing were all for the baby weight range of 24 to 32 pounds to correspond to the test apparatus described below.

The swim pants were placed onto an apparatus built to approximately simulate the shape and fit of a 24- to 32-pound baby. The "test baby" consisted of an oval-shaped wooden block approximately 6.0 inches long, 4.75 inches wide, and 1.5 inches thick. This wooden block was used to simulate the waist of the child. Two 500-mL plastic bottles (filled with water) were screwed into the bottom of the wooden block to act as the legs. Both the wooden block and the plastic bottles were then covered with neoprene (roughly one-half inch thick) to further simulate the actual fit of the swim pants around the skin on the leg and waist of a body. The neoprene material used was similar to that found in a wet suit or a kayaking skirt. After preliminary testing was performed, a second layer of neoprene was added to the wooden block to provide a tighter fit around the waist.

This simulation technique, by necessity of design, provided a "good-case" estimate of solids retention potential. The methods used probably represent the high end of actual solids retention, because the swim pants may have fit better around the neoprene-covered plastic bottles and wooden block than around an actual child. By necessity, the apparatus was designed to fit the products so that they would tend not to fall off during the active immersion testing.

Numerous attempts were made to include testing of regular disposable diapers as a control/comparison. Use of regular diapers in this way proved to be infeasible because when a typical waistline-taping tightness was used, the disposable diapers generally slid down and fell off the "test baby" within one to five minutes after the simulated play/activity movements were initiated. Even when the waist tape strips were put on much more tightly than any real baby would or could endure the glue on the waist tape became saturated with water, and the disposable diaper slipped off the "test baby" during the simulated water play period.

The authors chose not to use real babies, or real fecal material, because of the complications that would be involved in such tests. All of the solids used to test for fine-particulate retention by the swim pants were collected on a single day from a local construction site. This sampling site was chosen because it had been tilled and churned several times by backhoes and other construction equipment until a relatively homogeneous mixture of silt and clay resulted. The soils were collected from three different sections of the site to ensure randomness in the sampling procedure. The resulting mixture was a deeply red clay mixture with a few pebbles and soil clumps interspersed throughout. Hand sifting was performed at the site in order to remove any larger rocks, sticks, and debris. Once the solids were gathered, they were hand mixed to promote homogeneity. The damp soil was dried in a laboratory oven at 100[degrees]C for two hours (with one mixing after the first hour.) The samples were then removed from the oven and allowed to cool to room temperature. The uniformly dry soil was sifted to further ensure homogeneity: All of the soil was run through a one-quarter-inch galvanized mesh screen. The result was a dry, light, even soil void of any rocks or debris larger than one-quarter of an inch in diameter.

[ILLUSTRATION OMITTED]

The retention of solids by the swim pants was measured in terms of the parameters of turbidity and total suspended solids (TSS). These two tests were used to measure the amount of soil that was released by the swim pants over designated time intervals of 1.0 minutes, 5.0 minutes, 10.0 minutes, 20.0 minutes, and 30.0 minutes. Prior to each test, a 100-liter clear polyethylene drum was filled with tap water and tested for turbidity and TSS. This water provided the blank sample at time = 0.0 minutes and represented the baseline for each test. Then 100 grams of the dry, loose clay soil was placed into the swim pants (already on the baby apparatus). The "test baby" apparatus was immersed and pulled out of the water drum with a slow continuous motion resulting in about a 10-second cycle that included a vertical dunk followed by a lateral cross pattern.

The dunking and agitation was conducted for a total of 30 minutes with samples taken after 1.0 minutes, 5.0 minutes, 10.0 minutes, 20.0 minutes, and 30.0 minutes. At each designated time interval the timer was stopped, the "test baby" was removed from the 100-liter drum, and a PVC pipe was used to thoroughly mix solids and the water in the drum. The purpose of this step was to ensure that any settled particulate matter was accounted for in the measurement of turbidity and total suspended solids. At each designated time interval, a pre-labeled plastic laboratory bottle (250 mL for TSS and 125 mL for turbidity) was dipped into the center of the plastic drum and the water samples were extracted. The bottles used for sampling were tightly capped and placed in the refrigerator for later analysis. Once each dunking and agitation test was completed, the 100-liter drum was emptied and thoroughly rinsed with tap water to ensure that no particulate matter carried over to the next test.

Each sample was tested for TSS, which is the portion of total solids retained by a 0.8-micron filter, after the sample was shaken to ensure a complete mixing. This test measured the mass of soil that was released by the swim pants over the designated time intervals. The TSS analysis was conducted according to method "2540 D. Total Suspended Solids Dried at 103-105 Degrees Celsius" (American Public Health Association, 1998).

Turbidity measurements also were used to determine the relative solids retention abilities of the swim pants. Turbidity is the measure of suspended and colloidal matter in water. After the student researcher had shaken the sample to ensure a complete mixing, measurements were taken with a DRT-15CE turbidimeter manufactured by HF Scientific according to the "2130 B. Nephelometric Method" (American Public Health Association, 1998). The units used to measure turbidity are nephelometric turbidity units (NTUs).

The authors analyzed data through a repeated-measures analysis with PROC MIXED from the Statistical Analysis System (Litell, Millliden, Stroup, & Wolfinger, 1996). A no-intercept model was fitted for both turbidity and TSS with the following terms: time, time squared, time-with-brand interaction, and time-squared-with-brand interaction. The time-squared term was included to evaluate whether the relationship between time of agitation and turbidity or TSS is linear. The time-squared-with-brand-interaction term was included to evaluate whether some brands had a more nonlinear relationship than others. Assuming that not to be the case, the time-with-brand interaction would compare the performance of the three brands of diapers over time. (Brands with larger coefficient estimates for the time-with-brand-interaction term leach more solids over time.) A first-order autoregressive model was used for covariance structure of the individual diapers within brand despite the fact that sampling times were not equally spaced, because it fit most of the sample correlations for the data well.

[FIGURE 1 OMITTED]

Results and Discussion

From Table 1 it can be seen that both the time term, having a positive coefficient, and the time-squared term, having a negative coefficient, are highly statistically significant. This result indicates that TSS and turbidity are increasing as time increases and that the rate of increase slows at the longer times. The fact that neither the brand-with-time-squared term nor the brand-with-time term is statistically significant indicates that there is not evidence of any differences in the relationship between time and TSS or turbidity among the different brands studied.

From Figure 1 and Figure 2 it can be seen that the pattern of median concentrations versus time is similar for turbidity and total suspended solids. In both figures it appears that the IPlay swim pants have lower median turbidity and total suspended solids at the oneminute interval and that the Huggies swim pants produce lower values for 10 minutes, 20 minutes, and 30 minutes. The results of the mixed-model analysis, however, indicate that these apparent differences may be due to sampling variability as opposed to real differences in the performance of the different brands.

The percent retention of median TSS is shown in Figure 3. Percent retention is a linear function of TSS, and the authors derived it by converting the mg/L concentrations to total mass and accounting for the total amount of water used. The Huggies brand appears from the figure to have the highest percent retention of the three brands, but because percent retention is linearly related to TSS, there are no statistically significant differences among the brands for TSS results from the linear model. The mean percent solids retention of all three brands was greater than 98 percent even after 30 minutes of continuous dunking and agitation of the "test baby."

[FIGURE 2 OMITTED]

[FIGURE 3 OMITTED]

Conclusions

Overall, all three brands of swim pants performed well in retaining solids for 30 minutes of submerged agitation under the conditions of the tests. Each of the brands of diapers, however, released some solids within the first several minutes of agitation. While most of the diapers retained over 98 percent of the solids placed into them, none of the products completely eliminated solids loss in this study.

Implementing changes and re-evaluating the traditional safeguards at municipal facilities could greatly decrease the risk of fecal contamination. One way to accomplish such change could be to design and require the use of more effective swim pants. Manufacturers should tailor the swim pants to better fit the child; doing so would mean providing more sizing options (as IPlay has already done, with six sizes, compared with only three sizes offered by the other brands). Also, elastic bands and straps could be either custom fit or adjustable to achieve the highest possible level of fecal-matter retention.

Of necessity, these experiments had to produce a reasonably tight fit in order to conduct immersed agitation for 30 minutes without the swim pants falling off the simulated baby. This kind of fit clearly was not possible with traditional disposable diapers and, even when such diapers were taped with unrealistic tightness, the water immersion caused the tape to loosen and eventually come off. It was clear from visual inspection in these experiments that in real situations the IPlay brand, with its additional sizing options and tighter leg and waist elastic bands, would probably provide better waist and leg band fitting for more children than the other two brands, whose leg and waist bands were less elastic with fewer size choices. These preliminary tests indicate that swim diapers are relatively effective in retaining fine-particulate material as long as they fit tightly.

Children wearing conventional diapers should probably be restricted from municipal whole-body-contact recreation facilities. This public health problem can potentially affect any aquatic-recreation facility. Regulations should be established to ensure a high degree of protection from the type of contamination discussed in this report. Local community groups should tailor the regulations to their respective areas and their specific needs. Only through this type of comprehensive program can all swimming areas be made safe for young children and all swimmers to enjoy.

TABLE 1 Mixed-Model Analysis Results

                                    Log TSS        Log Turbidity
Term                            F-Value  P-Value  F-Value  P-Value

Time                            165.82   <.001    264.19   <.001
Time Squared                     69.08   <.001    128.13   <.001
Time-Brand Interaction            0.14    .887      1.68    .190
Time-Squared-Brand Interaction    0.01    .991      1.38    .255

REFERENCES

American Public Health Association. (1998). Standard methods for the examination of water and wastewater (20th ed.). Washington, DC: Author.

Centers for Disease Control and Prevention (2003). Healthy swimming2003.http://www.cdc.gov/health yswimming/. (22 May 2003).

Gilbert, L., & Blake, P. (1998). Out-break of Escherichia coli O157:H7 infections associated with a waterpark. Georgia Epidemiology Report, 14(7), 1-2.

Lee, S.H., Levy, D.A., Craun, G.F., Beach, M.J., & Calderon, R.L. (2002). Surveillance for waterborne-disease outbreaks United States, 1999-2000. Mortality and Morbidity Weekly Report, 51(SS-8), 1-28.

Littell, R.C., Milliden, G.A., Stroup, W.W., & Wolfinger, R.D. (1996). SAS system for mixed models. Cary, NC: SAS Institute.

Lohff, C.J., Nissen, G.M., Magnant, M.L., Quinlisk, M.P., Tiesdoetter, C.L., Kowalski, P.L., Buss, P.A., Link, T.A., Corrigan, M.R., Viner, J.P., Behnke, A.J., DeMartino, M.S., Houston, A.K. (2001). Shigellosis outbreak associated with an unchlorinated fill-and-drain wading pool--Iowa, 2001. Mortality and Morbidity Weekly Report, 50(37), 797-800.

Mead, P.S., Slutsker, L., Dietz, V., McCaig, L.F., Bresee, J.S., Shapiro, C., (1999). Food-related illness and death in the United States. Emerging Infectious Diseases, 5, 607-625.

Metz, P. (2000). Illinois swimming pool and bathing beach code. Illinois Department of Public Health.

Richard P. Maas, Ph.D.

Steven C. Patch, Ph.D.

Jacob F. Berkowitz

Holly D. Johnson

Corresponding Author: Steven C. Patch, Research Director, Environmental Quality Institute, University of North Carolia, Asheville, 212 Rhoades Hall, CPO #2331, One University Heights, Asheville, NC 28804-3299. E-mail: [e-mail].