Printer Friendly

New questions and insights into nitrate/nitrite and human health effects: a retrospective cohort study of private well users' immunological and wellness status.

Introduction

The health effects of nitrate/nitrite ingestion are thought to be well established (Bricker, Jefferson, & Mintz, 1983; Comly, 1945; Fan & Steinberg, 1996; Johnson et al., 1987; Kristen, 2001; Levallois & Phaneuf, 1994; Szponar & Kierzkowska, 1990; Walton, 1951; Ward et al., 2005). When nitrate or nitrite is used as a preservative in prepared food stuffs, ingested, and digested in the presence of nitrosable amines, nitrosamines are formed. Nitrosamines are linked to human esophageal and gastric cancers in a variety of epidemiological studies (Andreassi et al., 2001; Fewtrell, 2004; Gulis, Czompolyova, & Cerhan, 2002). When nitrate contaminates well water used for human consumption, it is ingested and transformed by bacteria in the gut to nitrite, which interacts readily with the hemoglobin molecule of the red blood cell. This process removes electrons from the ferrous iron in the normally functioning hemoglobin and creates a ferric iron methemoglobin state, which does not readily bind with oxygen, causing a biochemical anemia (Gatseva et al., 2000; Gebara & Goetting, 1994; Knobeloch, Salna, Hogan, Postle, & Anderson, 2000; Sadeq et al., 2008).

Since infants have small bodies but high water volume, they are very susceptible to the worst effects of acute nitrate toxicity due to the lower pH of the gut, an immature methemoglobin reductase system, high volume-to-surface-area ratios, and high ingestion rates of liquids. Biochemically induced "anemia," or infantile methemoglobinemia (iMHG), may result, which is not a loss of red blood cells but rather a lack of oxygen-carrying capacity (Abu Nasser, Ghbn, & Khoudary, 2007; Dagan, Zaltstein, & Gorodischer, 1988; Jolly, Monico, & McDevitt, 1995; Keating, Lell, Strauss, Zarkowsky, & Smith, 1973).

As a result of these well-documented health concerns, environmental and public health regulatory agencies around the world have made efforts to encourage the removal of excess nitrate/nitrite salts from foodstuffs and have set limits for the amount of nitrate/nitrite found in drinking water. For example, the U.S. Environmental Protection Agency (U.S. EPA) maximum contaminant level (MCL) is 10 parts per million (ppm) nitrate-nitrogen (nitrate-N) (Fewtrell, 2004; Hegesh & Shiloah, 1982; Lukens, 1987; Maloney, MacFarlane, & Rimsza, 1983).

Other researchers have correctly pointed out that nitrate is essential to human health. Nitrogen itself is a basic building block of protein; nitric oxide is utilized by both the neurological and immune systems to carry out vital bodily functions (Archer, 2002; Avery & Lhirondel, 2003; Billiar, Curran, Ferrarri, Williams, & Simmons, 1990; Duncan et al., 1997; Lhirondel & Lhirondel, 2002). Furthermore, some researchers and commentators have hypothesized that it is not the nitrate level alone that contributes to the acute health impacts of nitrate ingestion in infants but the comorbidity of infectious agents or inflammatory or allergic states that may potentiate the effects of nitrate for infants. Arguments follow that developed nations, where the water quality (biologically speaking) is much better, should be allowed leeway with respect to the current regulatory level, since comorbid factors such as diarrheal disease are less prevalent (Avery, 1999, 2001; Babbitt & Garrett, 2000; Gupta et al., 2007; Hanukoglu & Danon, 1996; Lebby, Roco, & Arcinue, 1993; Levine et al., 1998; May, 1985; Murray & Christie, 1993; Pollack & Pollack, 1994).

Dr. Zeman has worked with Romanian public health authorities since 1997 on environmental health issues including in-depth studies of rural water quality and iMHG cases. She published detailed exposure assessments in conjunction with case-control studies of iMHG that clearly indicated a dose-response relationship to infantile methemoglobinemia when highly contaminated water was ingested (Zeman, 2005). Dr. Zeman also reported in that same case-control study a bivariate association with iMHG and diarrheal disease. This association was not significant, however, under multivariate analysis in the Romanian groups. Rather, well water ingestion, nitrate level in the water, and early cessation of breast feeding were the most important variables in multifactor models. Intriguing questions remain regarding the bivariate relationship between iMHG and diarrheal disease and whether this relationship has any practical bearing on the setting of regulatory levels for nitrate in drinking water.

Dr. Zeman understands the alternative views about comorbidity concerning iMHG, diarrheal disease, and regulatory levels. Despite not finding any significant associations between presence of pathogenic microorganisms and iMHG after taking stool cultures from the Romanian case-control study participants, she worked again with the Romanian public health authorities and examined the same wells in Transylvania, this time for bacteria (fecal coliform, Streptococcal) and protozoa (Cryptosporidium, Giardia) that lead to diarrheal disease. iMHG cases were no more likely to have bacterial and protozoal contamination than were controls, although in both cases and controls the water contamination was quite high (Cryptosporidium mean = 3 cysts/L, range = 0-84; Giardia mean = 2 cysts/L, range = 0-36; fecal coliforms mean = 683/100 mL; fecal Streptococcal mean = 146/100 mL) (Zeman et al., 2005). Once again we were intrigued and sought to explain this apparent contradiction.

This result led to further research collaboration in the area of immunology as we hypothesized that perhaps the diarrheal association was due to confounding in the epidemiological study by a previously unstudied factor or factors, perhaps an immunotoxicological reaction. In vitro analysis of the proliferation rates and cytokine expression of lymphocytes from healthy volunteers exposed to sodium nitrate and nitrites (Ustyugova, Zeman, Dhanwada, & Beltz, 2002) was conducted. Both proliferation and cytokine expression were significantly impacted by the in-test-tube exposures; furthermore, these occurred at exposure levels that were in some cases below the MCL for nitrate. A variety of other researchers has also found relationships between nitrate/nitrite exposure and immunotoxicological effects (Abuharfeil, Jaran, Shabsough, & Darmani, 2001; Abuharfeil, Sarsour, & Hassuneh, 2001; Samlowski, Yim, & McGregor, 1998; Soderburg, 1994).

Could these immunotoxicological effects be detected in vivo in human populations as a result of environmental exposures? In order to try to answer this question, we conducted a pilot, retrospective cohort study of private well owners in Black Hawk County, Iowa, which included obtaining well water samples, health history data, and blood samples in which the participants' hemoglobin fractions and immunological parameters would be evaluated.

Methods

Our study utilized a retrospective cohort design targeting a population of 150 participants from [greater than or equal to] 1 year of age to [less than or equal to] 60 years of age. Individuals were recruited into cohorts based on the nitrate level of their drinking (well) water (high: <10.5 to >5 ppm; medium: <5 to >1 ppm; low: <1 ppm) and were then also divided into age groups of five years (1-60 years). Individuals with well water nitrate exposures exceeding 10.5 ppm nitrate-N who were less than one year or greater than 60 years of age and individuals with autoimmune diseases were excluded from the study. Individual recruitment began by locating private well owners through a state program for assisted well testing known as the "Grants to Counties" program. Additional recruitment outreach included a Web site, public service announcements, and targeted mailings with follow-up calls to zip codes known to contain a high proportion of private well owners.

Exclusion criteria included participants with any form of genetically mediated methemoglobinemia, including such conditions as hemoglobin M and genetically determined deficiencies in the b5 reductase enzyme. We also excluded people with any type of immune suppression or autoimmune disorder, such as rheumatoid arthritis; anyone undergoing kidney dialysis; and anyone experiencing work-related exposures to nitrate/nitrite-containing dyes, drugs, and so on. We excluded individuals with nitrate/nitrite levels in the target cohort categories who were treating their water with reverse osmosis filtration units or who were not consuming their well water; these individuals unaccounted for in the study design would confound the findings. Following fully informed consent of the participant (or the participant's legal guardian in the case of minors), a water sample was obtained and analyzed for assignment to cohort and an on-site interview was conducted by a public health nurse from the Black Hawk County Health Department.

The water analysis was completed on site in a portable laboratory using the chromoptropic acid method and the HACH DR/890 colorimeter. The interview involved answering a 10-page questionnaire consisting of a health status survey (including diarrheal disease history, general health history, medication history) and dietary nitrate intake/nitrate exposure history. The instrument was based on a field-tested survey instrument that was modified and derived from the U.S. Department of Defense health status questionnaire and the National Health and Nutrition Examination Survey dietary data questionnaire.

Once the fully informed consent was received, the on-site well assessment and interview were completed and the participant was scheduled to have a blood draw at one of two Wheaton Franciscan Health System sites. Following the blood draw, the laboratory notified University of Northern Iowa (UNI) research labs for pickup. UNI laboratory personnel were blinded as to the cohort status of the sample.

The sample was analyzed for hemoglobin fractions within two hours using multiple frequency co-oxiometry analysis. The AVOXimeter 4000 whole-blood oximeter is capable of multiple-band spectro-photometric analysis, which allows the researcher to quickly determine the oxyhemoglobin, deoxyhemoglobin, methemoglobin, and carboxyhemoglobin percentage and total hemoglobin (derived/calculated) in a whole blood sample. One to two mL of whole blood was injected into a disposable cuvette. The cuvette was inserted within 10 seconds of filling and read after the selection of sample type (i.e., whole blood) and readings were displayed as a digital printout with total hemoglobin (tHb), tHb as g/dL, and percentage of oxyhemoglobin, carboxyhemoglobin, and methemoglobin.

The remaining sample was analyzed within another one to two hours for immunological parameters. Lymphocytes and monocytes were isolated from other whole blood components using differential centrifugation applying the Ficoll-Hypaque technique (specific gravity = 1.077) followed by several washes with phosphate-buffered saline. The isolated peripheral blood mononuclear cells (PBMC) were then examined for viability (99% of cells were required to be viable to proceed), proliferation (vigor), and functional ability to produce normal biomolecules (cytokines) of immunity such as the Th1 cytokines interleukin-2 (IL-2), tumor necrosis factor-beta (TNF-[beta]), and the Th2/Treg cytokine interleukin-10 (IL-10). The proliferation assay involved cultivating the PBMC (2x[10.sup.5] PBMC/mL) in Roswell Park Memorial Institute (RPMI)-buffered growth medium and stimulating them with phytohemaglutinin (PHA) for 96 hours. This was followed by a four-hour pulse with 1 [micro]Ci (microcurie) [sup.3]H-thymidine (tritiated thymidine). Samples were harvested using a plate harvester and counts per minute (cpm) of thymidine incorporated into the DNA were determined using a beta-scintillation counter. Cell viability was determined by the 3-(4,5-dimethylthiamizol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) colormetric assay using a four-hour pulse with MTT following a 96-hour stimulation with PHA. The reaction was stopped with 0.4% HCl in isopropanol and read at 570 nm using a plate-reading spectrophotometer.

The production of the immunologically important IL-2, IL-10, and TNF-[beta] cytokines was accomplished using standard enzyme-linked immunosorbent assay (ELISA) techniques. The PBMC cells were first cultivated in RPMI and stimulated with PHA and the supernatants decanted. Captured antibodies were added to 96-microwell plates and incubated for 8-12 hours. Supernatant was added to the plates followed by detection antibodies plus enzyme and then substrate for 48 hours. Supernatants were collected and stored at -20[degrees]C/-4[degrees]F until analyzed using paired capture and detection antibodies with the reaction being stopped by carbonic acid and read at 450 nm using a plate-reading spectrophotometer. In every situation where analysis equipment was used in the field or laboratory, equipment was checked before each analysis for calibration and proper functioning.

The null hypothesis for our study required that cohorts experiencing varying exposures to nitrate/nitrite via the ingestion route will not differ from one another in their overall health status, diarrheal disease status, blood methemoglobin levels, indicators of lymphocyte competency (proliferation, viability, and select cytokine expression), or responses to a questionnaire consisting of a health status survey (including diarrheal disease history) and dietary nitrate intake/nitrate exposure history.

Statistical analyses of study results were accomplished using JMP 7.0 Statistical Discovery software, with a p = .05 level of significance for bivariate and multivariate analyses. Univariate (mean, mode, standard deviation, frequency distributions, etc.), bivariate (Chi-square, likelihood ratios, and Pearson's correlation coefficient), and multivariate analyses were completed (whole model tests such as nominal logistic fit, least square means, etc.), depending on the continuous or discrete nature of the variables.

Results

Descriptive Findings

Participants

While the original study design sought to include a large number of younger and older participant and as representative a sample of minorities as possible, these goals proved challenging in rural Iowa, which tends to be an older and more Caucasian population (Table 1). The sample consisted of 96% Caucasians and 4% racial or ethnic minorities (African-American, Hispanic, and Native American). The mean age of the participants was 45 years (SD = 15.2) with a range from six months to 60 years of age. The majority occupations were service industry workers (40%) and others (41%) (including retired, trucking, student, homemaker, and farming). Only 10% reported employment in industrial manufacturing. The average reported income of the sample was $41,400 (SD = $25,376).

Well Characteristics

The mean nitrate level of all well samples was 3.9 ppm nitrate-N (SD = 3.7 ppm, range = 0.25-10.5) (Table 2). The percentages of individuals having wells in the low (<1 ppm), medium (1-<5 ppm), and high (>5-<10.5 ppm) exposure cohorts were 34%, 28%, and 38%, respectively. Sixty-two percent of individuals had <5 ppm nitrate in their well water; 38% had >5 ppm. Seventy-four percent of the wells were of drilled construction, 15% were sandpoint, 4% were driven, and 7% were other/unknown. The average well depth was 129 feet (SD = 61.1 ft., range = 12-300) with an average well age of 34 years (SD = 20.9 yrs., range = 2 - 108).

Wellness Measures

The mean body mass index (BMI) for this population was 27.38 (SD = 7.12 BMI, range = 13.76-59.51) (Table 3). Thirty-two percent of the sample had a BMI in the normal range and 6% were underweight. The remaining 52% of the sample were classified as overweight (36%) or in one of three progressive categories of obesity: 5% in class I (BMI [less than or equal to] 34.99), 5% in class II (BMI [less than or equal to] 39.99), and 6% in class III (BMI [greater than or equal to] 40.00). These categories represent increasingly high BMI indexes with progressively worse categories of obesity; the third category represents the most overweight individuals. Ninety-four percent of the participants were nonsmokers who did not live with a smoker. Seventeen percent of the sample reported low or variable mood with 83% describing their mood as mostly good to excellent. Only 5% of the sample described their general health status as fair to poor, while the remaining 95% identified their general health status as good to excellent. Only 5% of the sample agreed that they became sick "more often than most people" they knew. Fifty-seven percent of the sample reported regular, weekly recreational activities producing an increase in heart rate, breathing, and light sweating of 10 minutes or more duration more than three times per week. Twenty percent of the sample reported the same physical activity three or fewer times per week and only 23% reported little or no physical recreational activity outside of work.

Hematological and Immunological Parameters Hemoglobin fractions, total hemoglobin values, and select immunological measures were obtained for all participants (Table 4). The mean hemoglobin was 14.07 g/100mL (SD = 2.08 g/100mL, range = 10.9-22.8). Mean oxyhemoglobin was 51.23% (SD = 18.27%, range = 23.3-99.9), which is low overall, even for venous blood (70% usual mean value). The percentage of total participants with a lower than expected oxyhemoglobin level was 83.3. This result had no correlation to gender, age, time of analysis, time from collection to analysis, or date of analysis. A slight correlation was found between lower oxyhemoblogin level and low reported levels of physical activity. Mean oxygen content was 9.56 mL/dL (SD = 3.42 mL/dL, range = 4.09-19.7); mean methemboglobin was 0.57% (SD = 0.37%, range = 0-1.65); and mean carboxyhemoglobin was 0.53% (SD = 1.35%, range = 0-7.4).

Immunological parameters are experimental values and have no established physiological normative values. The MTT stimulatory index and the [sup.3]H-thymidine stimulatory index are measures of lymphocyte health (Table 4). The MTT stimulatory index, a measure of mitochondrial activity in the lymphocytes and thus the number of viable cells, had a mean response ratio of 0.524 (SD = 0.156, range = 0.057-0.960). Dividing the stimulatory index response into three categories of low responder (0-1.2), medium responder (1.3-1.9), and high responder ([greater than or equal to]2) distributes 27.7%, 55.5%, and 16.8% of the sample, respectively. [sup.3]H-thymidine is measured as incorporation of radiolabelled thymidine into DNA during replication following stimulation of the lymphocytes. It is a measure of DNA synthesis activity and thus another measure of lymphocyte activity post challenge. The mean stimulatory index, the ratio of [sup.3]H-thymidine incorporated into stimulated versus unstimulated cells, was 21.5 (SD = 27.01, range = 0.60-144). Participants could be categorized as low (0-9.9), medium (10-40), and high (>41) responders for this parameter, which partitioned the sample into 50.5%, 33.3%, and 16.2%, respectively. This partitioning helped us to understand the pattern of response seen in the individual's PBMC due to their varying nitrate exposures and aided in statistical analysis of findings.

Production of cytokines by PBMC stimulated by PHA in vitro was analyzed by antigen-capture ELISA and expressed as picograms cytokine/milliliter (pcg/mL) (Table 4). Mean concentration of IL-2 was 586 pcg/mL (SD = 952 pcg/mL, range = 36-4,992). The sample was partitioned into low expressers (0-100), 30.3%; medium expressers (101-1,000), 46.1%; and high expressers (>1,001), 23.7%. Mean concentration of TNF-P was 3,486 pcg/mL (SD = 3,970 pcg/mL, range = 70-14,932). The sample was partitioned into low expressers (0-1,000), 46.7%; medium expressers (1,001-8,000), 33.7%; and high expressers (>8,000), 19.6%. Mean concentration of IL-10 was 1,079 pcg/mL (SD = 1,029 pcg/mL, range = 86-5,789). The sample was again partitioned into low expressers (0-500), 40.2%; medium expressers (501-2,000), 45.7%; and high expressers (>2,001), 14.1%. Partitioning the response patterns of the PBMC in regard to their cytokine and interleukin expression aided in understanding individual responses to nitrate exposure and in our statistical analysis of findings.

Physical Ailments

Determination of physical complaints and diagnoses was based on subject self-reports (Table 5). Forty-four and a half percent of the sample reported past health complaints of any kind and 77.8% of the sample reported current health complaints of any kind when asked to reflect in detail over the previous 30 days. Of the wide variety of health complaints reported in Table 5, the top five, in order of frequency, along with the most frequently reported complaint by specific category were allergy (sneezing, nasal congestion, watery eyes), 45.6%; stomach/intestinal complaints (heartburn and reflux), 23.4%; heart/circulatory problems (hypercholesterolemia), 22.8%; bone/muscle/nerve (arthritis), 22.1%; and behavioral/emotional (depression), 18%.

Bivariate Findings

Wellness Measures

Significant relationships to wellness measures were revealed by bivariate analyses and included relationships between nitrate exposure and BMI (calculated), recreational activity (self-reported), perceived health (self-reported), and perceptions of susceptibility to illness (self-reported) (Table 6). A statistically significant bivariate fit between high nitrate exposures and overweight, all obesity classes, and underweight was found (f = 4.77, p = .0312). Individuals exposed to higher levels of nitrate also reported less recreational activity/exercise (Log fit, Chi-square = 31.19, p = .0001). Using the one-way analysis of variance, relationships were explored between exposure to higher levels of nitrates via drinking water sources, poorer perceived health (f = 3.02, p = .0320) and getting sicker more easily (f = 3.87, p = .0051) than others known to the participant, with significant findings among these variables.

Hematological and Immunological Measures

A directly proportionate relationship was seen in our study between methemoglobin level in the blood and increased levels of nitrate ingestion through drinking water sources (Table 6). While this was seen only as a trend if data were analyzed as continuous values using bivariate fit, the trend was toward significance (f = 3.50, p = .06); recoding the data by >5 and <5 ppm nitrate in drinking water sources and using logistic fit illustrated a significant finding with higher levels of methemoglobin as percentage of hemoglobin fraction found in venous blood with increasing exposure to nitrate (Chi-square = 4.12, p = .0423) although in no case was the level above physiologically normal values. A persistent relationship existed between TNF-[beta] (high expression) and increasing levels of nitrate exposure as a continuous value using bivariate fit methods (f = 3.76, p = .05). This trend strengthened when the data was examined by high, medium, and low level of exposure with the logistic fit, whole model test (Chi-square = 4.46, p = .0348). When data for IL-10 (high expression) was examined following the exclusion of two outliers, a trend toward significance was seen between groups using well water with >5 and <5 ppm of nitrate (f = 1.94, p = .1667).

Physical Ailments

Increased complaints of stomach/intestinal difficulties (majority heartburn/reflux > 50%, f = 5.274, p = .0231) and bone, muscle, nerve complaints (majority osteoarthritis = 47%, f = 6.0533, p = .0150) were found with increased exposure to nitrate in drinking water (Table 6).

Whole Model Tests

Using nominal logistic fit, we found that both nitrate-N (ppm) and age were predictive of gastrointestinal complaints (p = .0109, .0032, respectively) (Table 7). Using the same test we found that both nitrate-N (ppm) and age were predictive of bone/joint health complaints (p = .0109, .0032, respectively). Using stepwise fit and least square means we found that TNF-[beta] (increased production following challenge) was positively associated with >5 ppm nitrate-N in drinking water (p = .0450), nitrate-N (as a continuous value in ppm, p = .0450), and the intensity of recreational activities over the past 30 days (p = .0522). Finally, the fit least squares test indicated a positive relationship between IL-10 (increased production following challenge) and BMI (p = .0413) and bone/joint health complaints (p = .0546). While the data on specific bone/joint and gastrointestinal complaint was recoded by specific complaint type in an effort to evaluate specific complaints using whole model/multivariate approaches, the small number of data points did not allow for comparison at this level of detail.

Conclusion

Positive associations were found between reliance on higher nitrate water sources and self-reports/perceptions of being less healthy than others. This was also true in self-evaluations of being less active recreationally than lower nitrate water consumers and of lower and higher than normal body mass indexes. The exact nature of these relationships from a causal time sequence perspective should be explored in further studies that couple detailed exposure assessment and these variables. For example, it is known that a large portion of the nitrate exposure in an adult's diet comes from vegetable consumption and the majority of water consumed is thought to be consumed in the home environment (73%-75%; Hopkins & Ellis, 1980). The water consumption in an adult's diet (and the water's nitrate level), however, is in addition to this solid food intake and to some degree (depending on choice of drinking water source) more controllable; exploring these relationships in future studies in addition to wellness and activity measures would further clarify these relationships.

Further, higher nitrate consumers reported more complaints of gastrointestinal illnesses, bone/joint disorders, and exhibited a high production of the TNF-[beta] cytokine. IL-10 cytokine production was positively associated with lower BMI and with complaints of bone/joint disorders. The associations with nitrate, health complaint, and TNF-[beta] held constant when considered along with important demographic factors such as age when run as a multivariate analysis. Age, for example, is known to have relationships or possible relationships to these variables, and while age was noted as a cosignificant factor in whole model tests, this may be associated with years of exposure and should be considered as a possibility.

The epidemiological literature illustrates a history of nitrite exposures and gastrointestinal and esophageal cancers associated with those exposures; thus, the association with gastrointestinal disorders/complaints is not as surprising as a positive correlation with complaints of bone/joint disorders (Fan & Steinberg, 1996).

The research literature in the area of osteoarthritis and related disorders indicates reports dating back to 1997, when it was proposed that pro-inflammatory cytokines induce nitric oxide production through the inducible nitric oxide synthase (iNOS) pathway and that this plays a role in the etiology of both rheumatoid arthritis and osteoarthritis (Grabowski et al., 1997). A study in 2001 examined this hypothesis in animal models (mice) and it was supported with findings indicating that the nitric oxide had an impact on both osteoblast and osteoclast activity (Armour et al., 2001). Serum nitrate and nitrite levels were examined in individuals with osteoarthritis, rheumatoid arthritis, and ankylosing spondylitis and found to be elevated compared to controls in a 2002 study (Ersoy et al., 2002). A study of over 5,000 elderly white women who took nitrate medication for chest pain found twice the risk of osteoarthritis of the hip when compared to controls who did not take nitrate medication (Lane et al., 2004). An examination of human cartilage tissue from osteoarthritic joints found elevated levels of interleukin-1 (IL-1) and tumor necrosis factor-alpha (TNF-[alpha]) and indications of the iNOS pathway being induced in response to these factors, which led to elevated nitric oxide levels in the cartilage as a strong predictor of osteoarthritis pathology (Grabowski et al., 1997; Vuolteenaho, Moilanen, Knowles, & Moilanen, 2007). Interestingly, a study examining the effectiveness of the herb Ajuga decumbens found that its effectiveness could be linked to the inhibition of the iNOS pathway and a decrease in nitric oxide levels in the joint tissue (Yuka, Fukaya, Imai, & Yamakuni, 2008).

Our study is the first known documented instance of in vivo exposures from environmental sources of nitrate-N (water supply) being positively associated with complaints of bone/joint disorders, or with altered ex vivo production of TNF-[beta] or IL-10 cytokines by the human immune system following such in vivo nitrate exposure. TNF-[beta] is a Th1 cytokine, which tends to be stimulatory for the immune system and may produce inflammatory reactions. IL-10 is a cytokine produced by stimulated Th2 and Treg lymphocytes and monocytes that tends to moderate the immune response and protect against inflammatory diseases, including various forms of arthritis.

Significantly, IL-10 decreases production of pro-inflammatory cytokines, such as TNF-[alpha], IL-1, interleukin-6, and the molecules involved in joint damage, such as nitric oxide, collagenase, and gelatinase (Pawlik, Kurzawski, Szklarz, Hercyzynska, & Drozdzik, 2005). Levels of IL-10 and Treg lymphocytes are increased by urocortin, an immunomodulator used in the treatment of arthritis. Antibodies against IL-10 reversed the therapeutic effects of urocortin, indicating the importance of this cytokine in protection against arthritis (Gonzalez-Rey, Chorny, Varela, O'Valle, & Delgado, 2007). The increased levels of IL-10 found in our study may be due to a protective system by which the body attempts to decrease the actions of pro-inflammatory cytokines. The general survey nature of our study, however, did not allow for multivariate analysis of specific bone/joint disorders, nor was the original survey instrument designed for this purpose. Additional, more detailed studies with smaller sample sizes and a restricted hypothesis are indicated based on our findings.

Our previous in vitro work indicated that immunological perturbations were likely; this result was confirmed in TNF-[beta] and a trend was identified with IL-10. Previous work in our laboratories as well as that of other researchers confirms the relationship between nitrate/nitrite ingestion and interactions with immune factors (Ustyugova et al., 2002; Wyatt et al., 2006). This is another important area for further research as these impacts are noted with MCL-compliant water sources. Additionally other pro-inflammatory cytokines are likely of importance, particularly in the association found between nitrate ingestion and bone/joint complaints, those being IL-1, IL-6, and TNF-[alpha] (Kopp et al., 2005). These should be investigated further using a larger sample size with a narrowed hypothesis. The role of another Treg cytokine, transforming growth factor-beta, should also be explored in the future. Preliminary data in our laboratory suggests that in vitro exposure of normal human lymphocytes to nitrate and nitrite increases the production of this cytokine, which has been correlated with protection against damage due to arthritis.

Our study in and of itself is not conclusive nor exhaustive about health factors related to nitrate ingestion. It does indicate the need for further research, however, when considered in light of other scientific research and when the findings are associated in populations where exposure levels are at or under the current MCL. In particular, our study is a pilot study and could be improved upon by a number of additional, larger studies with narrowed hypotheses and a broader number of water contaminants, including metals, organics, and pesticides.

In concluding this discussion, it is important to note that our study was accomplished as a joint effort of environmental health practitioners and university research scientists. This collaboration provided scientists in the university setting with invaluable community "groundedness" and with a wealth of practical field experience that would not have otherwise been possible. Additionally, it tied the field practice into the generation of new knowledge.

Humanity is now converting gaseous nitrogen from the atmosphere into bioavailable, water-soluble nitrogen and introducing it into terrestrial ecosystems more than twice the naturally fixed amount of nitrogen; this trend has over the last decade continued to accelerate and increase. The importance of understanding the human health ramifications of these anthropogenic changes is more important than ever (Criss & Davisson, 2004; Fields, 2004; Mannion, 1998; Mitchell, 2006). As we are now living in a world where the biogeochemical cycles for the major macronutrients (nitrogen cycle, carbon cycle) are experiencing unprecedented fluxes, perturbations, and "overloads," the multiple skills and knowledge brought to bear in our study hold particular importance for the recognition and study of non-cancer endpoints that may be related to these perturbations. This collaborative and applied interdisciplinary research model leads us to look at regulatory levels from new perspectives and becomes critical to a vibrant practice of environmental health.

Acknowledgements: The authors wish to acknowledge the National Institutes of Environmental Health Science (NIEHS) R-15 grant for support of this work, and the administrative and laboratory staff of the Wheaton Franciscan Health System Hospitals of Cedar Falls and Waterloo, Iowa, for their phlebotomy contributions.

References

Abuharfeil, N., Jaran, A., Shabsough, B., & Darmani, H. (2001). Effects of sodium nitrate on natural killer cells isolated from human peripheral blood. Archives of Toxicology, 75(5), 291-296.

Abuharfeil, N., Sarsour, E., & Hassuneh, M. (2001). The effect of sodium nitrite on some parameters of the immune system. Food and Chemical Toxicology, 39(2), 119-124.

Abu Nasser, A., Ghbn, N., & Khoudary, R. (2007). Relation of nitrate contamination of groundwater with methaemoglobin level among infants in Gaza. East Mediterranean Health Journal, 13(5), 994-1004.

Andreassi, M.G., Picano, E., Del Ry, S., Botto, N., Colombo, M.G., & Giannessi, D. (2001). Chronic long-term nitrate therapy: Possible cytogenic effect in humans? Mutagenesis, 16(6), 517-521.

Archer, D.L. (2002). Evidence that ingested nitrate and nitrite are beneficial to health. Journal of Food Protection, 65(5), 872-875.

Armour, K.J., Armour, K.E., van't Hof, R.J., Reid, D.M., Wei, X., Liew, FY., & Ralston, S.H. (2001). Activation of the inducible nitric oxide synthase pathway contributes to inflammation-induced osteoporosis by suppressing bone formation and causing osteoblast apoptosis. Arthritis and Rheumatism, 44(12), 2790-2796.

Avery, A.A. (1999). Infantile methemoglobinemia: Reexamining the role of drinking water nitrates. Environmental Health Perspectives, 107(7), 583-586.

Avery, A.A. (2001). Cause of methemoglobinemia: Illness versus nitrate exposure. Environmental Health Perspectives, 109(1), A12-A14.

Avery, A., & Lhirondel, J. (2003). Nitrate and methemoglobinemia. Environmental Health Perspectives, 111(3), A142-A143.

Babbitt, C.J., & Garrett, J.S. (2000). Diarrhea and methemoglobinemia in an infant. Pediatrics Emergency Care, 16(6), 416-417.

Billiar, T.R., Curran, R.D., Ferrarri, F.K., Williams, D.L., & Simmons, R.L. (1990). Kupffer cell: Hepatocyte coculture release nitric oxide in response to bacterial endotoxin. Journal of Surgical Research, 48(4), 349-353.

Bricker, T., Jefferson, L.S., & Mintz, A.A. (1983). Methemoglobinemia in infants with enteritis. Journal of Pediatrics, 102(1), 349-353.

Comly, H.H. (1945). Cyanosis in infants caused by nitrates in well water. Journal of the American Medical Association, 129(2), 112-116.

Criss, R.E., & Davisson, M.L. (2004). Fertilizers, water quality, and human health. Environmental Health Perspectives, 112(10), A536.

Dagan, R., Zaltzstein, E., & Gorodischer, R. (1988). Methaemoglobinaemia in young infants with diarrhoea. European Journal of Pediatrics, 147(1), 87-89.

Duncan, C., Li, H., Dykhuizen, R., Frazer, R., Johnston, P., Macknight, G., Smith, L., Lamza, K., McKenzie, H., Batt, L., Kelly, D., Golden, M., Benjamin, N., & Leifert, C. (1997). Protection against oral and gastrointestinal diseases: Importance of dietary nitrate intake, oral nitrate reduction and enterosalivary nitrate circulation. Comparative Biochemistry and Physiology, 118(4), 939-948.

Ersoy, Y., Ozerol, E., Baysal, O., Temel, I., MacWalter, R.S., Meral, U., & Altay, Z.E. (2002). Serum nitrate and nitrite levels in patients with rheumatoid arthritis, ankylosing spondylitis, and osteoarthritis. Annals of the Rheumatic Diseases, 61(1), 76-78.

Fan, A.M., & Steinberg, V.E. (1996). Health implications of nitrate and nitrite in drinking water: An update on methemoglobinemia occurrence and reproductive and developmental toxicity. Regulatory Toxicology and Pharmacology, 23(1), 35-43.

Fewtrell, L. (2004). Drinking-water nitrate, methemoglobinemia, and global burden of disease: A discussion. Environmental Health Perspectives, 112(14), 1371-1374.

Fields, S. (2004). Global nitrogen: Cycling out of control. Environmental Health Perspectives, 112(10), A556-A563.

Gatseva, PV, Mardirosian, Z.H., Popova, E.J., Iskrenova, E.S., Vladeva, S.V., & Pavlova, K.I. (2000). Evaluation of health hazards in children from regions with nitrate pollution. Folia Medica, 42(1), 19-22.

Gebara, B.M., & Goetting, M.G. (1994). Life-threatening methemoglobinemia in infants with diarrhea and acidosis. Clinical Pediatrics, 33(6), 370-373.

Gonzalez-Rey, E., Chorny, A., Varela, N., O'Valle, F, & Delgado, M. (2007). Therapeutic effect of urocorin on collagen-induced arthritis by down-regulation of inflammatory and Th1 responses and induction of regulatory T cells. Arthritis and Rheumatism, 56(2), 531-543.

Grabowski, P.S., Wright, P.K., van't Hof, R.J., Helfrich, M.H., Oshima, H., & Ralston, S.H. (1997). Immunolocalization of inducible nitric oxide synthase in synovium and cartilage in rheumatoid arthritis and osteoarthritis. British Journal of Rheumatology, 36(6), 651-655.

Gulis, G., Czompolyova, M., & Cerhan, J.R. (2002). An ecologic study of nitrate in municipal drinking water and cancer incidence in Trnava District, Slovakia. Environmental Research, 88(3), 182-187.

Gupta, S.K., Gupta, R.C., Gupta, A.B., Seth, A.K., Bassin, J.K., Gupta, A., & Sharma, M.L. (2007). Recurrent diarrhea in children living in areas with high levels of nitrate in drinking water. Archives of Environmental Health, 56(4), 369-373.

Hanukoglu, A., & Danon, P.N. (1996). Endogenous methemoglobinemia associate with diarrheal disease in infancy. Journal of Pediatric Gastroenterology and Nutrition, 23(1), 1-7.

Hegesh, E., & Shiloah, J. (1982). Blood nitrates and infantile methemoglobinemia. Clinica Chimica Acta, 125(2), 107-115.

Hopkins, S.M., & Ellis, J.C. (1980). Drinking water consumption in Great Britain: A survey of drinking habits with special reference to tap-water based beverages (Technical Report No. 137). Wiltshire, Great Britain: Water Research Centre.

Johnson, C.J., Bonrud, P.A., Dosch, T.L., Kilness, A.W., Senger, K.A., Busch, D.C., & Meyer, M.R. (1987). Fatal outcome of methemoglobinemia in an infant. Journal of the American Medical Association, 257(20), 2796-2797.

Jolly, B.T., Monico, E.P., & McDevitt, B. (1995). Methemoglobinemia in an infant: Case report and review of the literature. Pediatric Emergency Care, 11(5), 294-297.

Keating, J.P., Lell, M.E., Strauss, A.W., Zarkowsky, H., & Smith, G.E. (1973). Infantile methemoglobinemia caused by carrot juice. New England Journal of Medicine, 288(16), 824-826.

Knobeloch, L., Salna, B., Hogan, A., Postle, J., & Anderson, H. (2000). Blue babies and nitrate-contaminated well water. Environmental Health Perspectives, 108(7), 675.

Kopp, S., Alstergren, P., Ernestam, S., Nordahl, S., Morin, P., & Bratt, J. (2005). Reduction of temporomandibular joint pain after treatment with a combination of methotrexate and inflixima c45b is associated with changes in synovial fluid and plasma cytokines in rheumatoid arthritis. Cells, Tissues, Organs, 180(1), 22-30.

Kristen, K. (2001). Nitrates linked to bladder cancer. Environmental Science and Technology, 35(13), 279A-280A.

Lane, N.E., Williams, E.N., Hung, Y.Y., Hochberg, M.C., Cummings, S.R., & Nevitt, M.C. (2004). Association of nitrate use with risk of new radiographic features of hip osteoarthritis in elderly white women: The study of osteoporotic fractures. Arthritis and Rheumatism, 49(6), 752-758.

Lebby, T., Roco, J.J., & Arcinue, E.L. (1993). Infantile methemoglobinemia associated with acute diarrheal illness. American Journal of Emergency Medicine, 11 (5), 471-472.

Levallois, P., & Phaneuf, D. (1994). Contamination of drinking water by nitrates: Analysis of health risks [Article in French]. Canadian Journal of Public Health, 85(3), 192-196.

Levine, J.J., Pettei, M.J., Valderrama, E., Gold, D.M., Kessler, B.H., & Tratchtman, H. (1998). Nitric oxide and inflammatory bowel disease: Evidence for local intestinal production in children with active colonic disease. Journal of Pediatric Gastroenterology and Nutrition, 26(1), 34-38.

L'hirondel, J., & L'hirondel, J.L. (2002). Nitrate and man: Toxic, harmless, or beneficial? Wallingford, Oxfordshire, United Kingdom: CABI Publishing.

Lukens, J.N. (1987). The legacy of well-water methemoglobinemia. Journal of the American Medical Association, 257(20), 2793-2795.

Maloney, W.J., MacFarlane, M.T., & Rimsza, M.E. (1983). Methemoglobinemia in an infant. Arizona Medicine, 40(10), 700-702.

Mannion, A.M. (1998). Global environmental change: The causes and consequences of disruption to biogeochemical cycles. The Geographical Journal, 164(2), 168-182.

May, R.B. (1985). An infant with sepsis and methemoglobinemia. Journal of Emergency Medicine, 3(4), 261-264.

Mitchell, S. (2006). Nitrate, agriculture and the environment. The Geographical Journal, 172(4), 351-352.

Murray, K.F, & Christie, D.L. (1993). Dietary protein intolerance in infants with transient methemoglobinemia and diarrhea. Journal of Pediatrics, 122(1), 90-92.

Pawlik, A., Kurzawski, M., Szklarz, B.G., Hercyzynska, M., & Drozdzik, M. (2005). Interleukin-10 promoter polymorphism in patients with rheumatoid arthritis. Clinical Rheumatology, 24(5), 480-484.

Pollack, E.S., & Pollack, C.V (1994). Incidence of subclinical methemoglobinemia in infants with diarrhea. Annals of Emergency Medicine, 24(4), 652-656.

Sadeq, M., Moe, C.L., Attarassi, B., Cherkaoui, I., Elaouad, R., & Idrissi, L. (2008). Drinking water nitrate and prevalence of methemoglobinemia among infants and children aged 1-7 years in Moroccan areas. International Journal of Hygiene and Environmental Health, 211(5-6), 546-554.

Samlowski, W.E., Yim, C.Y., & McGregor, J.R. (1998). Nitric oxide exposure inhibits induction of lymphokine-activated killer cells by inducing precursor apoptosis. Nitric Oxide, 2(1), 45-56.

Soderburg, L.S. (1994). T cell functions are impaired by inhaled isobutyl nitrite through a T-independent mechanism. Toxicology Letters, 70(3), 319-329.

Szponar, L., & Kierzkowska, E. (1990). Nitrates and nitrites in environment and their effect on human health [Article in Polish]. Postepy Higieny i Medycyny Doswiadczalnej, 44(4-6), 327-350.

Ustyugova, I.V, Zeman, C., Dhanwada, K., & Beltz, L.A. (2002). Nitrates/nitrites alter human lymphocyte proliferation and cytokine production. Archives of Environmental Contamination Toxicology, 43(3), 270-276.

Vuolteenaho, K., Moilanen, T., Knowles, R.G., & Moilanen, E. (2007). The role of nitric oxide in osteoarthritis. Scandinavian Journal of Rheumatology, 36(4), 247-258.

Walton, G. (1951). Survey of literature relating to infant methemoglobinemia due to nitrate-contaminated water. American Journal of Public Health, 41(8 Pt. 1), 986-996.

Ward, M.H., deKok, T.M., Levallois, P., Brender, J., Gulis, G., Nolan, B.T., VanDerslice, J., & International Society for Environmental Epidemiology. (2005). Workgroup report: Drinking-water nitrate and health--Recent findings and research needs. Environmental Health Perspectives, 113(11), 1607-1614.

Wyatt, N., Kelly, C., Fontana, V., Merlo, D.F, Whitelaw, D., & Anderson, D. (2006). The responses of lymphocytes from Asian and Caucasian diabetic patients and non-diabetics to hydrogen peroxide and sodium nitrite in the Comet assay. Mutation Research, 609(2), 154-164.

Yuka, O., Fukaya, Y., Imai, S., & Yamakuni, T. (2008). Beneficial effects of Ajuga decumbens on osteoporosis and arthritis. Biological & Pharmaceutical Bulletin, 31(6), 1199-1204.

Zeman, C. (2005). Infant methemoglobinemia: Causative factors. Environmental Health Perspectives, 113(12), A805-A806.

Zeman, C.L., Seuleanu, I., Sinca, M., Sinca, A., Moga, D., & Vlad, M. (2005). Environmental illness may have contributed to the origins of Transylvanian vampire myths. International Journal of Global Health, 3(2), 29-39.

Catherine Zeman, MS, PhD

Lisa Beltz, MS, PhD

Mark Linda, MS

Jean Maddux

Diane Depken, Edd

Jeff Orr, MA

Patricia Theran, MA

Corresponding Author: Catherine Zeman, Associate Professor and Director, Environmental Health, University of Northern Iowa, Health Division/RRTTC, WRC 207, Cedar Falls, IA 50614-0241. E-mail: catherine.zeman@uni.edu.
TABLE 1
Participant Demographics

Demographic                 Mean       SD       Range      %

Age (years) (n = 147)        45       15.2      0.5-60    N/A
Gender (n = 145)             N/A       N/A       N/A
  Female                                                  58
  Male                                                    42
Ethnicity (n = 146)          N/A       N/A       N/A
  Caucasian                                               96
  Other                                                    4
Occupation (n = 165)         N/A       N/A       N/A
  Service industry                                        40
  Industry                                                10
  Education                                                9
  Other                                                   41
Income (n = 101)           $41,400   $25,376   $10,000-   N/A
                                               $107,000

TABLE 2
Well Characteristics

Demographic                       Mean    SD      Range      %

Cohort (ppm (a)) (n = 164)        N/A    N/A       N/A
  High (<10.5)                                              38
  Medium (>5)                                               28
  Low (<1)                                                  34
  % > 5                                                     38
  % < 5                                                     62
Nitrate level (ppm) (n = 149)
  Whole sample                    3.9    3.7    0.25-10.5   N/A
Well type (n = 149)               N/A    N/A       N/A
  Drilled                                                   74
  Sandpoint                                                 15
  Driven                                                     4
  Other                                                      7
Well depth (feet) (n = 142)       129    61.1    12-300     N/A
Well age (years) (n = 128)         34    20.9     2-108     N/A

(a) Parts per million.

TABLE 3 Wellness Measures

Demographic                           Mean     SD       Range       %

BMI (a) (n = 103)                     27.38   7.12   13.76-59.51   N/A

BMI classification (n = 103)           N/A    N/A        N/A
  Underweight                                                       6
  Overweight                                                       36
  Obese 1                                                           5
  Obese 2                                                           5
  Obese 3                                                           6
  Normal                                                           32

Mood, spirits (n = 149)                N/A    N/A        N/A
  Excellent                                                        22
  Very good                                                        33
  Mostly good                                                      28
  Up/down                                                          16
  Low                                                               1

Health status (n = 147)                N/A    N/A        N/A
  Excellent                                                        31
  Very good                                                        47
  Good                                                             17
  Fair                                                              5

Activity level (per wk.) (n = 149)     N/A    N/A        N/A
  >3                                                               57
  <3                                                               20
  Unable                                                            2
  No                                                               21

Sick easier than others? (n = 148)     N/A    N/A        N/A
  Definitely false                                                 57
  Mostly false                                                     34
  Unknown                                                           4
  Mostly true                                                       4
  Definitely true                                                  <1

Smoker? (n = 107)                      N/A    N/A        N/A
  No                                                               94
  Yes                                                               6

(a) Body mass index.

TABLE 4
Hematological and Immunological Parameters

Demographic                               Mean     SD       Range

Hemoglobin (tot. g/100mL) (n = 138)       14.07   2.08    10.9-22.8

Oxyhemoglobin (%) (n = 138)               51.23   18.27   23.3-99.9

Sample with low oxyhemoglobin              --      --        --
level (%) ([dagger]) (n = 138)

Oxygen content (mL/dL) (n = 138)          9.56    3.42    4.09-19.7

Methemoglobin (%) (n = 138)               0.57    0.37     0-1.65

Carboxyhemoglobin (%) (n = 138)           0.53    1.35      0-7.4

MTT (mitochondrial activity/optical       0.524   0.156    0.057-
density; stimulatory index) (n = 122)                       0.960

[sup.3]H, tritiated thymidine             21.5    27.01   0.60-144
(DNA synthesis activity; stimulatory
index) (n = 100)

IL-2, interleukin 2                        586     952     36-4992
(concentration; pcg/mL) (n = 76)

TNF-[beta], tumor necrosis factor         3486    3970    70-14,932
(concentration; pcg/mL) (n = 92)

IL-10, interleukin 10                     1079    1029     86-5789
(concentration; pcg/mL) (n = 92)

Demographic                                    %

Hemoglobin (tot. g/100mL) (n = 138)            --

Oxyhemoglobin (%) (n = 138)                    --

Sample with low oxyhemoglobin                 83.3
level (%) ([dagger]) (n = 138)

Oxygen content (mL/dL) (n = 138)               --

Methemoglobin (%) (n = 138)                    --

Carboxyhemoglobin (%) (n = 138)                --

MTT (mitochondrial activity/optical       LR (a): 27.7
density; stimulatory index) (n = 122)     MR (a): 55.5
                                          HR (a): 16.8

[sup.3]H, tritiated thymidine               LR: 50.5
(DNA synthesis activity; stimulatory        MR: 33.3
index) (n = 100)                            HR: 16.2

IL-2, interleukin 2                       LE (b): 30.3
(concentration; pcg/mL) (n = 76)          ME (b): 461
                                          HE (b): 23.7

TNF-[beta], tumor necrosis factor           LE: 46.7
(concentration; pcg/mL) (n = 92)            ME: 33.7
                                            HE: 19.6

IL-10, interleukin 10                       LE: 40.2
(concentration; pcg/mL) (n = 92)            ME: 45.7
                                            HE: 14.1

(a) LR: low responder, MR: medium responder, HR: high responder.

(b) LE: low expresser, ME: medium expresser, HE: high expresser.

([dagger]) Low oxyhemoglobin < 70% for venous blood draw.

TABLE 5
Physical Ailments

Questionnaire Ailment Categories        Specific Ailments (a)      %

Past health complaints (any kind)                 --              44.5
Current health complaints (any kind)              --              77.8
Prematurity                             Subject born premature    1.3
Seizures                                          --               2
Diabetes                                          --              5.3
Tumor/cancer                             Skin cancers (46.6%)      10
Kidney/bladder                          Kidney stones (33.3%)      16
Heart                                 Hypercholesterolemia (44%)  22.8
Lung/breathing                               Asthma (50%)         17.5
Blood                                             --               8
Skin                                        Eczema (58.3%)        10.7
Stomach/intestinal                          Reflux (58.6%)        23.4
Diarrhea/loose stools                             --               6
Weight change                                     --               2
Bone/muscle/nerve                         Arthritis (46.9%)       22.1
Eye/vision                             Corrective lens (87.5%)     66
Ear/hearing                              Hearing loss (72.7%)     8.7
Behavioral/emotional                      Depression (70.8%)       18
Developmental delay                      Subject experienced       <1
                                         delayed development
Syndromes                                         --               <1
Allergy                                           --              45.6

(a) Most frequently reported problem listed with percentage if
greater than 10% of sample population.

TABLE 6
Various Bivariate Analyses

Variable                           Test            Test Statistic

Wellness Measures

X: nitrate ppm (a)            Bivariate fit,       F-ratio: 4.7745
Y: BMI (b) (overweight,         ANOVA (c)
underweight, and all
obesity classes)

X: nitrate ppm                Logistic fit,       Chi-square: 31.19
Y: recreational activity        Chi-square
(less exercise)

X: perceived health           One-way ANOVA        F-ratio: 3.0153
(poorer)
Y: nitrate ppm

X: get sick easier than       One-way ANOVA        F-ratio: 3.8691
others (yes)
Y: nitrate ppm

Hematological and Immunological Measures

X: < or > 5 ppm nitrate       One-way ANOVA        F-ratio: 4.0602
Y: methemoglobin %

X: ppm                        Bivariate fit,       F-ratio: 3.7574
X: cohort H, M, L (d)      ANOVA, logistic fit,      Chi-square:
Y: TNF-[beta] (e) (high      whole model test         4.455752
expression)

X: < or > 5 ppm nitrate       One-way ANOVA+       F-ratio: 1.9441
Y: IL-10 (f) (high
expression)

Physical Ailment Reports

X: stomach/ intestinal        One-way ANOVA        F-ratio: 5.2736
complaints
Y: nitrate ppm

X: bone/muscle/nerve          One-way ANOVA        F-ratio: 6.0533
complaints
Y: nitrate ppm

Variable                       p-Value

Wellness Measures

X: nitrate ppm (a)         .0312 *
Y: BMI (b) (overweight,
underweight, and all
obesity classes)

X: nitrate ppm             .0001 *
Y: recreational activity
(less exercise)

X: perceived health        .0320 *
(poorer)
Y: nitrate ppm

X: get sick easier than    .0051 *
others (yes)
Y: nitrate ppm

Hematological and Immunological Measures

X: < or > 5 ppm nitrate    .0459 *
Y: methemoglobin %

X: ppm                     .0557
X: cohort H, M, L (d)      .0348 *
Y: TNF-[beta] (e) (high
expression)

X: < or > 5 ppm nitrate    .1667 ([dagger])
Y: IL-10 (f) (high
expression)

Physical Ailment Reports

X: stomach/ intestinal     .0231 *
complaints
Y: nitrate ppm

X: bone/muscle/nerve       .0150 *
complaints
Y: nitrate ppm

(a) Parts per million.

(b) Body mass index.

(c) Analysis of variance.

(d) Cohorts were designated high, medium, and low (H, M, L)
based on the nitrate level of their well water.

(e) Tumor necrosis factor-beta.

(f) Th2/Treg cytokine interleukin-10.

* p-value of .05 required for statistical significance.

([dagger]) IL-10 levels increased as the level of nitrate
exposure in drinking water increased; despite the lack of
statistical significance, this trend is reported for future
studies.

TABLE 7
Whole Model Tests

Test              Response      Parameters
                  Variable

Nominal           Gastro-       Age, gender, occupation,
logistic fit     intestinal     smoking, nitrate-N (ppm (a))
                    (GI)
                  problems

Nominal          Bone/joint     Age, gender, occupation,
logistic fit      problems      smoking, nitrate-N (ppm)

Stepwise       TNF-[beta] (b)   Gender, occupation,
fit                             income, BMI (c), smoking,
                                rec. last 30 days,
                                nitrate-N (ppm),
                                nitrate-N <,>5 ppm,
                                GI prob., bone/joint
                                prob.

Least            TNF-[beta]     Income, BMI, smoking,
square                          rec. last 30 days, nitrate-N
means                           (ppm), GI prob., bone/joint
                                prob.

Fit least        IL-10 (d)      Gender, occupation,
squares                         income, BMI, smoking,
                                rec. last 30 days, nitrate-N
                                (ppm), nitrate-N <,>5
                                ppm, GI prob., bone/joint
                                prob.

Test           Significant Parameters   [R.sup.2]    F-Ratio

Nominal         Nitrate-N p = .0109)        --         --
logistic fit       Age p = .0032)

Nominal        Nitrate-N (p = .0109)        --         --
logistic fit      Age (p = .0032)

Stepwise         Nitrate-N <,>5 ppm      .09 for       --
fit                  (p = .045)         parameter,
                                         .20 for
                                          model

Least          Nitrate-N (p = .0450)     .23 for      4.31
square                                    model       3.21
means            Rec. last 30 days
                    (p = .0522)

Fit least         BMI (p = .0413)        .25 for      4.36
squares                                   model       3.85
                  Bone/joint prob.
                    (p = .0546)

Test           Chi-Square   Odds Ratio   Confidence
                                          Interval

Nominal           6.48        0.265      0.74-0.09
logistic fit      8.69        0.046      0.40-0.002

Nominal           6.74        0.239      0.08-0.70
logistic fit     17.61        0.002        0.07-
                                          0.00002

Stepwise           --           --           --
fit

Least              --           --           --
square
means

Fit least          --           --           --
squares

(a) Parts per million.

(b) Tumor necrosis factor-beta.

(c) Body mass index.

(d) Th2/Treg cytokine interleukin-10.
COPYRIGHT 2011 National Environmental Health Association
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2011 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:ADVANCEMENT OF THE SCIENCE
Author:Zeman, Catherine; Beltz, Lisa; Linda, Mark; Maddux, Jean; Depken, Diane; Orr, Jeff; Theran, Patricia
Publication:Journal of Environmental Health
Article Type:Report
Geographic Code:1USA
Date:Nov 1, 2011
Words:8119
Previous Article:Pay it forward ...
Next Article:Arsenic levels in the soil and risk of birth defects: a population-based case-control study using GIS technology.
Topics:

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