The Zuni Kidney Project: A Collaborative Approach to an Epidemic of Kidney Disease
Christine A. Stidley*,
Vallabh O. Shah,
Marina Scavini,,
Andrew S. Narva¶,
David Kessler¶,
Arlene Bobelu,
Jean W. MacCluer**,
Thomas K. Welty and
Philip G. Zager
Departments of *Family and Community Medicine, and Internal Medicine, University of New Mexico; Dialysis Clinic Inc., Albuquerque, New Mexico; and ¶Kidney Disease Program, Indian Health Service, Albuquerque, New Mexico; **Southwest Foundation for Biomedical Research, San Antonio, Texas; and Istituto Scientifico H San Raffaele, Milan, Italy.
Correspondence to Dr. Philip G. Zager, University of New Mexico, Department of Internal Medicine, Nephrology ACCS, Albuquerque, NM 87131-5271. Phone: 505-247-4044; Fax: 505-247-1297;
ABSTRACT. There is an epidemic of renal disease among the ZuniIndians. In contrast to most other American-Indian communities,the epidemic of renal disease among the Zuni Indians reflectshigh rates of diabetic and nondiabetic renal disease. Almostevery Zuni Indian has a relative with end-stage renal disease.This epidemic offers a unique opportunity to advance our understandingof the risk factors for the susceptibility and/or progressionof renal disease. Thus, Zuni Tribal leaders formed a researchpartnership with the University of New Mexico Health SciencesCenter, Indian Health Service, Southwest Foundation for BiomedicalResearch and Dialysis Clinic Inc., to establish the Zuni KidneyProject (ZKP). The ZKP conducted a population-based, cross-sectionalsurvey of the Zuni Pueblo. Age and gender distributions amongsurvey participants were similar to those of the eligible Zunipopulation. Among diabetics the prevalence (95% confidence interval)of incipient albuminuria (IA) was 32.3% (25.1, 39.5) in womenand 36.1% (24.7, 47.5) in men. The prevalence of IA among nondiabeticswas 9.3% (6.9, 11.7) in women and 12.2% (9.7, 14.7) in men.Among diabetics, the prevalence of overt albuminuria (OA) was17.7% (11.9, 23.5) in women and 20.8% (11.4, 30.2) in men. Amongnondiabetics, OA was present in 1.2% (0.3, 2.1) of women and2.3% (1.1, 3.5) of men. Although IA and OA were each more commonamong diabetics, the majority of participants with albuminuriawere nondiabetics. Hematuria was common among both diabeticsand nondiabetics. Among diabetics, the crude prevalence of hematuriawas similar among men and women. Among nondiabetics, however,hematuria was more common among women. Diabetes and obesitywere more common among women than men. In contrast, hypertensionand hypercholesterolemia were more common among men than women.The ZKP is incorporating these preliminary data into planningfor the development and implementation of primary and secondaryprevention programs. E-mail: pzag@unm.edu
In the United States, minorities carry a disproportionate burdenof renal disease. In 2000, point prevalence (per million population[pmp]) of ESRD, adjusted for age and gender, was higher in African-(4240 pmp) and Native Americans (3287 pmp) than in EuropeanAmericans (943 pmp) (1). The higher prevalences among AfricanAmericans and Native Americans, compared with European Americans,reflect higher incidence rates and longer survival. The prevalenceof ESRD, adjusted for age and gender, among the Zuni Indiansis 17,400 pmp (2). This is 4.1- and 18.5-fold higher than thatof African and European Americans, respectively. Moreover, theprevalence of ESRD among the Zuni Indians is 5.3-fold higherthan that in the composite Native-American population (1).
The causes of renal disease vary significantly among American-Indiantribes (3). The prevalence of type 2 diabetes mellitus is veryhigh among American Indians. In most American-Indian communities,the vast majority of renal disease is attributable to diabetes.The Pima Indians are experiencing a well recognized epidemicof renal disease due to diabetic nephropathy (4). It is lesswidely appreciated that the Zuni Indians are also experiencingan epidemic of renal disease. In contrast to the Pima Indians,the epidemic of renal disease among the Zuni Indians reflectshigh rates of both diabetic and nondiabetic renal disease. Thelatter is most commonly a mesangiopathic glomerulopathy (MesGN) (5), often IgA nephropathy. Pasinski and Pasinski reportedthat between 1973 and 1983, chronic glomerulonephritis and diabeticnephropathy, respectively, accounted for 40% and 24% of ESRDamong the Zuni (6). Hoy et al. (7) and Hughson et al. (8) reportedthat a high proportion of renal biopsy specimens from Zuni Indiansshowed Mes GN, which was frequently associated with IgA positivityon immunofluorescence and electron dense deposits on electronmicroscopy.
Study Population
The Zuni Pueblo is located in a rural portion of western NewMexico. The Zuni Indians have lived in the present Pueblo foralmost 400 yr. They differ genetically and culturally from theNavajos and other American-Indian tribes. Low rates of emigrationand immigration indicate that the community has been relativelyendogamous. The 2000 Tribal Census recorded over 2000 householdsand 10,228 members (9). The median age was 26 yr. Only 8% ofthe population was over 60 yr of age. The community retainsa strong traditional component. Ninety percent of residentsspeak the Zuni language. The community is economically disadvantagedand approximately half the population lives below the povertyline. The economy is heavily dependent upon jewelry making andother artistic crafts. However, the community established theinfrastructure (e.g., electricity, water, and sewer-systems)necessary to maintain public health.
Establishment of the Zuni Kidney Project
Almost every Zuni Indian has a relative with ESRD. This ledto broad-based community support for renal research on primaryand secondary prevention. The Zuni tribal leaders recognizedthat an innovative approach was necessary to obtain the informationnecessary to design and implement effective primary and secondaryprevention programs. Thus, they established collaborative researchpartnerships with the Indian Health Service (IHS), Universityof New Mexico Health Sciences Center (UNMHSC), Southwest Foundationfor Biomedical Research (SFBR), Dialysis Clinic Inc. (DCI),and National Institutes of Health (NIH). These collaborationsculminated in the establishment of the Zuni Kidney Project (ZKP).The ZKP is currently exploring the hypothesis that both geneticand environmental factors modulate the risk for susceptibilityand/or progression of diabetic and nondiabetic renal disease.The results of this research will be incorporated into the planningof primary and secondary prevention strategies.
The specific aims of the ZKP are to: (1) educate the communityabout kidney disease; (2) conduct a population-based, cross-sectionalsurvey (PBCSS) to obtain precise estimates of the prevalenceof kidney disease and putative risk factors; (3) conduct a case-controlstudy to identify environmental and vocational risk factorsfor kidney disease; (4) conduct longitudinal studies of cohortswith normal, incipient, and overt albuminuria defined by urinaryalbumin creatinine ratio (UACR) expressed as true ratio; (5)conduct a study of extended Zuni families to identify risk factorsfor renal disease and intermediate phenotypes; and (6) developstrategies to prevent the occurrence and progression of renaldisease. This paper describes the development of the ZKP, educationalprograms, methodology, and preliminary results from the PBCSS,and plans for the development and implementation of primaryand secondary prevention programs.
The UNMHSC Human Research Review Committee, IHS InstitutionalReview Board, and the Zuni Tribal Council approved the project.Informed consent was obtained from each adult participant andfrom a parent or legal guardian for each participant less than18 yr of age. The Zuni language is not a written language, sothe consent form was written in English. Study staff translatedthe consent form verbally for those tribal members who couldnot read English.
Education
Because the ZKP grew from the community’s vision of aunique research partnership, we developed effective programsto educate the community about the diagnosis, treatment, andprevention of kidney disease. The project’s study coordinator(AB) is a professional educator, who previously served the communityas the director of the science curriculum for the Zuni publicschools. Under her leadership, the ZKP developed a wide varietyof educational programs. These programs used different venuesto disseminate information, including school programs, communityhealth fairs, public service announcements on the local radiostation, and articles in the Shiwi Messenger, the pueblo’slocal newspaper. These efforts heightened community awarenessand stimulated interest in acquiring additional informationabout kidney disease. Many tribal members visited the ZKP officeto talk with project staff and to acquire additional printededucational material.
Population-Based Cross-Sectional Survey (PBCSS)
The PBCSS is the foundation of the ZKP. The specific aims ofthe PBCSS include to: (1) estimate the prevalence of renal disease;(2) assess potential risk factors for renal disease; and (3)identify participants for planned case-control, longitudinal-cohort,and family studies to identify environmental, familial, andgenetic risk factors for the susceptibility and/or progressionof renal disease. All residents 5 yr of age wereeligible for participation in the PBCSS. The ZKP recognizedthe importance of random selection of participants. However,random sampling of individuals was not feasible. The Zuni TribalOffice provided a list of houses (n = 2201). Household clustersof 12 neighboring houses were identified (n = 221). These werevisited in a sequence established using a random number generator.Return visits were scheduled to maximize ascertainment. Interviewersvaried the times of visits to maximize participation. Two staffmembers visited each household. During the visit, study staffexplained the PBCSS, answered questions, and obtained written,informed consent. The questionnaire was administered and bloodand urine samples were obtained. BP (BP) was measured accordingto the American Heart Association criteria. Height and weightwere measured. Participants received compensation for theirtime and effort. Results from urine and blood tests were givento participants and referrals made when appropriate. Venousblood was drawn in labeled Vacutainer tubes. Clean catch urinesamples were collected in sterile containers. A chemistry profile,serum creatinine, complete blood cell count, HbA1c, urine albumin,and urine creatinine were measured (2). Quality control proceduresfor the collection and processing of laboratory samples havebeen previously described (2).
Data Management
All data obtained from the PBCSS were entered into an ACCESS(Microsoft, Seattle, WA) database. To monitor contacts withcommunity members, a tracking system was developed. Householdswere assigned a unique identifier. All contact dates, namesof participants and canvassers, and results of each visit wererecorded. We recorded laboratory reference numbers and datesfor enrollment, procurement of laboratory samples, notificationof participants of results, and physician referrals. Trackingreports checked the accuracy of questionnaires and laboratorysamples. The data management system had restricted user access.
Statistical Analysis
Data were analyzed using SAS (SAS Institute, Cary, NC) and SUDAAN(RTI, Research Triangle Park, NC). Standard statistical methodswere not appropriate due to the complex sampling scheme. SUDAAN,a statistical software package for analysis of complex samples,and SAS were used to obtain prevalence estimates and standarderrors. SAS was also used to fit generalized estimating equationsto model disease status. These models can be considered an extensionof logistic regression models that accommodate the samplingprocedure. Prevalences of selected variables were age- and gender-adjustedto the 2000 Zuni Tribal Census (9).
PBCSS Questionnaire
A questionnaire, based on the Behavioral Risk Factor SurveillanceSurvey and the Strong Heart Study questionnaires (2), was developedto obtain information pertinent to renal disease and relatedrisk factors. The instrument was refined through discussionswith the tribal leadership and the community to ensure compliancewith Zuni social norms and cultural customs. The questionnairecontains sections on household, demographics, medical and physicalhistory, social history, and exposure to potential risk factors(tobacco use, lack of exercise, alcohol consumption, drinkingwater supply, nonalcoholic drinks), family structure, and healthhistory. Reproducibility and validity of the questionnaire wereestablished by assessing responses on duplicate administrationsand comparing responses with medical records, respectively.
Classification of Participants
To control costs, enhance efficiency, and maximize recruitment,we used recently validated epidemiologic screening tools (e.g.,single spot urine samples for UACR and HbA1c to screen for diabetes)(10–11). Albuminuria was classified according to AmericanDiabetes Association (ADA) guidelines as normal (UACR <0.3),incipient (IA) (0.03 UACR < 0.3), or overt (OA) (0.03)(12). We considered using UACR gender-specific cut-points (13).However, among ZKP participants, the ratio of the mean urinarycreatinine for women to men was 0.90. Thus, we classified participantsaccording to the ADA guidelines (12), which do not recommendgender-specific cut-points. Participants were classified asdiabetic if they had a prior history of diabetes, a random glucose200 mg/dl (14), or HbA1c >7.0% (11). Participantsabove 20 yr of age were classified as overweight if their bodymass index (BMI) was 25 and <30, and as obeseif BMI 30 (15). Among participants 5 to 19 yrof age, BMI percentiles were obtained from gender-specific growthcharts (16). Participants were classified as overweight if theirBMI was the 85th and <95th percentiles andobese if their BMI was 95th percentile (17). Participants18 yr of age were classified as hypertensive ifthey had a prior history of hypertension, a systolic BP 140mmHg, or a diastolic BP 90 mmHg (18). Children(<18 yr) were classified as hypertensive if they had a priorhistory of hypertension, or had systolic or diastolic BP BORDER="0">95th percentile for age and height obtained fromgender-specific charts (19). Hypercholesterolemia was definedas a total serum cholesterol 170 mg/dl and BORDER="0">200 mg/dl for those <20 and 20 yrof age, respectively.
The preliminary results presented here are based on participantsenrolled in the PBCSS. We compared the demographics of participantsto the Zuni population using the 1990 (20) and 2000 censuses(9). The sample and the censuses were similar with respect togender, age, and education, indicating that study participantswere likely representative of the eligible population (2). Theprevalence of albuminuria among survey participants, stratifiedby age, gender, and diabetes status, is shown in Table 1. Theprevalence (95% confidence interval) of IA among diabetics was32.3% (25.1, 39.5) in female and 36.1% (24.7, 47.5) in maleparticipants. The prevalence of IA among nondiabetics was 9.3%(6.9, 11.7) in women and 12.2% (9.7, 14.7) in male participants.The prevalence of OA was also higher among diabetics than nondiabetics.Although the prevalence of albuminuria was higher among diabeticsversus nondiabetics, the majority of participants with albuminuriawere nondiabetics.
Table 1. Prevalence (%) of incipient and overt albuminuria, stratified by age, gender, and diabetesa
The prevalences of putative risk factors for albuminuria (e.g.,overweight, obesity, high cholesterol, and hypertension stratifiedby age and gender) are shown in Table 2. Both diabetes and obesitywere more common among female than male Zuni Indians. In contrast,hypertension and hypercholesterolemia were more common amongmale Zuni Indians.
Table 2. Age-adjusted prevalence (%) of overweight, obesity, high cholesterol, and hypertension, stratified by age and gendera
The crude prevalence of hematuria (dipstick trace),stratified by age, gender, and diabetes, is shown (Table 3).Hematuria was common among both diabetics and nondiabetics.Among diabetics, the crude prevalence of hematuria was similarin male and female participants. Among nondiabetics, however,the crude prevalence of hematuria was higher among female thanmale participants.
The epidemic of renal disease in the Zuni Indians is uniqueamong American-Indian communities because of the high prevalenceof nondiabetic renal disease. Specifically, this study demonstratesthat the prevalence of albuminuria is higher among diabeticsthan nondiabetics. However, because the number of nondiabeticparticipants greatly exceeded the number of diabetic participants,the majority of Zuni Indians with albuminuria did not meet thecriteria for type 2 diabetes mellitus. However, because theprevalence of diabetes is very high and increases with advancingage, we anticipate that a significant proportion of nondiabeticparticipants with albuminuria may eventually develop type 2diabetes mellitus. Putative risk factors for albuminuria (e.g.,diabetes, obesity, and hypertension were common among the ZuniIndians).
In concert with the wishes of the community, renal biopsieswere not performed for research purposes during this study.Previous reports have demonstrated that both diabetic nephropathyand Mes GN occur frequently among the Zuni Indians (21). IgAnephropathy is the most common form of Mes GN among the ZuniIndians. Diabetic nephropathy and Mes GN may coexist in thesame person (8). The high rates of hematuria observed amongdiabetics and nondiabetics are consistent with the hypothesisthat Mes GN may frequently occur among diabetic and nondiabeticZuni Indians.
The identification of risk factors and people with incipientrenal disease will facilitate the development and implementationof primary and secondary prevention strategies. The educationprograms will increase the community’s knowledge of renaldisease and motivate behaviors to decrease the risk for thedevelopment and progression of renal disease. The ZKP will workclosely with the Zuni Wellness Program and the Zuni DiabetesProgram to reduce the prevalence of diabetes, a major risk factorfor renal disease. The ZKP will develop a Hypertension ControlProgram to decrease the risk for progression of renal disease.The case-control study will allow the identification of vocationaland environmental risk factors. This will enable the ZKP tosuggest implementation of changes in the work place and homeenvironments to reduce the risk of kidney disease. The plannedlongitudinal cohort studies will allow the ZKP to identify andmodify risk factors for progression and improve secondary preventionprograms. The planned study of extended families will identifygenetic risk factors. This will facilitate identification ofindividuals at high risk who can then be targeted for earlytherapeutic intervention.
Acknowledgments
We acknowledge the contributions of the Zuni Governor, TribalCouncil, and Zuni Study staff. We thank the Zuni people fortheir gracious and continuous support. This research was fundedin part by the National Institutes of Health (DK 49347, 07/01/1997to 06/30/2002), University of New Mexico Clinical Research Center(NIH NCRR GCRC Grant 5MO1 RR OO997), and Dialysis Clinic, Inc.The opinions expressed in this paper are those of the authorsand do not necessarily reflect those of the Indian Health Service.
US Renal Data System: USRDS 2001 Annual Data Report: Atlas of End-Stage Renal Disease in the United States, Bethesda MD, NIH, National Institute of Diabetes and Digestive and Kidney Diseases, 2001
Stidley CA, Shah VO, Narva AS, Dalton D, MacCluer JW, Bobelu A, Scavini M, Welty TK, Zager PG: A population-based, cross-sectional survey of the Zuni Pueblo: A collaborative approach to an epidemic of kidney disease. Am J Kidney Dis 39: 358–368, 2002[Medline]
Robbins DC, Knowler WC, Lee ET, Yeh J, Go OT, Welty T, Fabsitz R, Howard BV: Regional differences in albuminuria among American-Indians: An epidemic of renal disease. Kidney Int 49: 557–563, 1996[Medline]
Nelson RG, Morgenstern H, Bennett PH: An epidemic of proteinuria in Pima Indians with type 2 diabetes mellitus. Kidney Int 54: 2081–2088, 1998[CrossRef][Medline]
Narva AS, Beaver S, Blackman D, Zack M, Freeman W: Descriptive study of renal disease at Pueblo of Zuni. Am J Kidney Dis 20: A8, 1992
Pasinski R, Pasinski M: End-stage renal disease among the Zuni Indians: 1973–1983. Arch Intern Med 147: 1093–1096, 1987[Abstract]
Hoy WE, Megill DM, Hughson MD: Epidemic renal disease of unknown etiology in the Zuni Indians. Am J Kidney Dis 9: 485–496, 1987[Medline]
Hughson MD, Megill DM, Smith SM, Tung KS, Miller G, Hoy WE: Mesangiopathic glomerulonephritis in Zuni (New Mexico) Indians. Arch Pathol Lab Med 113: 148–157, 1989[Medline]
Office of the Zuni Tribal Census: Tribal Census 2000. Zuni Pueblo, 2001
Schwab SJ, Christensen RL, Dougherty K, Klahr S: Quantitation of proteinuria by use of protein-to-creatinine ratios in single urine samples. Arch Intern Med 147: 943–944, 1987[Abstract]
Rohlfing CL, Little RR, Wiedmeyer HM, England JD, Madsen R, Harris MI, Flegal KM, Eberhardt MS, Goldstein DE: Use of GHb (HbA1c) in screening for undiagnosed diabetes in the U.S. population. Diabetes Care 23: 187, 191, 2000
American Diabetes Association: Clinical practice recommendations 2001. Diabetes Care 24: S69–S72, 2001
Warram JH, Gearin G, Laffel L, Krolewski AS: Effect of duration of type I diabetes on the prevalence of stages of diabetic nephropathy defined by urinary albumin/creatinine ratio. J Am Soc Nephrol 7: 930–937, 1996[Abstract]
American Diabetes Association: Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 20: 1183–1197, 1997[Medline]
Garrow JS, Webster J: Quetelet’s index (W/H2) as a measure of fatness. Int J Obes 9: 147–153, 1985[Medline]
Kuczmarski RJ, Ogden CL, Grummer-Strawn LM, Flegal KM, Guo SS, Wei R, Mei Z, Curtin LR, Roche AF, Johnson CL: CDC growth charts: United States. Adv Data 1: 27, 2000
Barlow SE, Dietz WH: Obesity evaluation and treatment: Expert Committee recommendations. The Maternal and Child Health Bureau, Health Resources and Services Administration and the Department of Health and Human Services. Pediatrics 102: E29, 1998
Joint National Committee on Prevention, Detection, Evaluation, And Treatment of High Blood Pressure: Sixth Report. Arch Intern Med 157: 2413–2446, 1997[Abstract]
Kuczmarski RJ, Ogden CL, Grummer-Strawn LM, Flegal KM, Guo SS, Wei R, Mei Z, Curtin LR, Roche AF, Johnson CL: CDC growth charts: United States. Adv Data 1: 27, 2000
Bureau of the Census: General Population Characteristics, American Indian and Alaska Native Areas,CP-1–1A, US government publication, 1990
Smith SM, Tung KS: Incidence of IgA-related nephritides in American-Indians in New Mexico. Hum Pathol 16: 181–184, 1985[CrossRef][Medline]
,
Top
Abstract
Introduction
5 yr of age were eligible for participation in the PBCSS. The ZKP recognized the importance of random selection of participants. However, random sampling of individuals was not feasible. The Zuni Tribal Office provided a list of houses (n = 2201). Household clusters of 
12 neighboring houses were identified (n = 221). These were visited in a sequence established using a random number generator. Return visits were scheduled to maximize ascertainment. Interviewers varied the times of visits to maximize participation. Two staff members visited each household. During the visit, study staff explained the PBCSS, answered questions, and obtained written, informed consent. The questionnaire was administered and blood and urine samples were obtained. BP (BP) was measured according to the American Heart Association criteria. Height and weight were measured. Participants received compensation for their time and effort. Results from urine and blood tests were given to participants and referrals made when appropriate. Venous blood was drawn in labeled Vacutainer tubes. Clean catch urine samples were collected in sterile containers. A chemistry profile, serum creatinine, complete blood cell count, HbA1c, urine albumin, and urine creatinine were measured (2). Quality control procedures for the collection and processing of laboratory samples have been previously described (2). 
