Secondary Prevention of Renal and Cardiovascular Disease: Results of a Renal and Cardiovascular Treatment Program in an Australian Aboriginal Community
Wendy E. Hoy,
Zhiqiang Wang,
Philip R. A. Baker and
Angela M. Kelly*
*Menzies School of Health Research, Darwin, Northern Territory;Centre for Chronic Disease, University of Queensland, Australia; Queensland Health, Brisbane, Australia.
Correspondence to Dr. Wendy E Hoy, Center for Chronic Disease, University of Queensland, Brisbane, Queensland, Australia. Phone: 61-07-334-64809; Fax: 61-07-334-64812;
ABSTRACT. Australian Aborigines are experiencing an epidemicof renal and cardiovascular disease. In late 1995 we introduceda treatment program into the Tiwi community, which has a three-to fivefold increase in death rates and a recent annual incidenceof treated ESRD of 2760 per million. Eligible for treatmentwere people with hypertension, diabetics with micro or overtalbuminuria, and all people with overt albuminuria. Treatmentcentered around use of perindopril (Coversyl, Servier), withother agents added to reach BP goals; attempts to control glucoseand lipid levels; and health education. Thirty percent of theadult population, or 267 people, were enrolled, with a meanfollow up of 3.39 yr. Clinical parameters were followed every6 mo, and rates of terminal endpoints were compared with thoseof 327 historical controls matched for baseline disease severity,followed in the pretreatment program era. There was a dramaticreduction in BP in the treatment group, which was sustainedthrough 3 yr of treatment. Albuminuria and GFR stabilized orimproved. Rates of natural deaths were reduced by an estimated50% (P = 0.012); renal deaths were reduced by 57% (P = 0.038);and nonrenal deaths by 46% (P = 0.085). Survival benefit wassuggested at all levels of overt albuminuria, and regardlessof diabetes status, baseline BP, or prior administration ofangiotensin converting enzyme inhibitors (ACEI). No significantbenefit was apparent among people without overt albuminuria,nor among those with GFR less than 60 ml/min. An estimated 13renal deaths and 10 nonrenal deaths were prevented, with thenumber-needed-to-treat to avoid one terminal event of only 11.6.Falling deaths and renal failure in the whole community supportthese estimates. The program was extremely cost-effective. Programslike this should be introduced to all high-risk communitiesas a matter of urgency. E-mail: wendy@kdrp.org
Aborigines living in remote Australia have high rates of all-causemortality and of cardiovascular deaths and renal failure (ESRD)(1,2), with the incidence of treated ESRD exceeding 1000 permillion in some regions. Treatment costs are spiraling out ofcontrol. From 1996 to 1998, the annualized treatment costs foran Aboriginal person on hemodialysis in the top end of the NorthernTerritory (NT) was estimated at $112,169: $71,000 for dialysistreatments alone and $41,169 for intercurrent hospitalizations(3).
These problems have been especially serious in the Tiwi Islands(population about 1800, 50% age 20+ yr), with high rates ofpremature adult death, and an incidence of treated ESRD reaching2760 per million in the mid-1990s (2). In the 1990s, 25% ofnatural deaths in Tiwi adults were renal deaths, and 42% hada primary or underlying cardiovascular cause (4).
In a community-wide screening program, we showed that the underlyingrenal disease is marked by albuminuria, which is multideterminant;it increases with age so that 55% of all adults are affected,and is inversely correlated with GFR (5,6). Albuminuria marksall of the future risk for renal deaths, but also strongly predictsnonrenal deaths, both cardiovascular and nonrenal noncardiovascular(4).
Use of anti-hypertensive drugs, including angiotensin convertingenzyme inhibitors (ACEI), was gradually increasing in the Tiwicommunity through the early 1990s, but systematic managementof the burden of disease exposed by our studies was beyond thecapabilities of the health services in place. We therefore introducedin November 1995 a program to alter the renal and cardiovasculardisease profile.
We previously described results of this program to December1998 (7). We now summarize the results through June 30, 2000.The endpoint data are described in more detail in a separatemanuscript (8).
The program operated out of one large community clinic on BathurstIsland and two smaller clinics on Melville Island.
The albumin creatinine ratio (ACR, g/mol) on a random urinespecimen was used as the marker of renal disease and its severity(6). The following categories were used: <3.4, normal; 3.4to 33, microalbuminuria; and 34+, overt albuminuria. Overt albuminuriawas further categorized thus: 34 to 99, moderate; 100 to 199heavy; 200+ intense. Serum creatinine levels were measured,and GFR was estimated by the Cockcroft-Gault formula (9). Inmid-1996, we changed laboratories that assayed urine ACR andserum creatinine levels, although the techniques remained thesame. Duplicate assays were performed on 47 urine specimensfor ACR and on 57 serum specimens for creatinine assays. Regressionequations indicated that the correlations were excellent (R20.98 and 0.99, respectively), with levels in the second laboratoryslightly higher than the first. All ACR and creatinine datawere analyzed on the unadjusted levels, as well as the "lab-adjusted"levels.
People eligible for treatment were those with confirmed hypertension(140/90 on two occasions), all diabetics withconfirmed pathologic albuminuria (ACR, 3.4 g/mol),and all persons with confirmed overt albuminuria (ACR 34+),regardless of BP or diabetes. Exclusions were people who werepregnant, breast feeding, or known to be allergic to ACEI. Ingeneral, sicker people were enrolled earlier in the program;these included several people with progressive renal insufficiencywho were failing their current regimens, which usually includedACEI. The final "treatment group" consisted only of people whowere prescribed perindopril for at least 1 mo. All analysesinvolving the treatment cohort were conducted on an "intention-to-treat"basis, which included all patients enrolled, regardless of theirsubsequent compliance with medication.
Medical treatment centered around the use of the long-actingangiotensin-converting enzyme inhibitor, perindopril (Coversyl,Servier), with BP goals of below 130/85 in the first 2 yr ofthe program, and below 125/75 in the last 2 yr (10). The intendedminimum perindopril dose was 4 mg, but the dose was increasedto 8 mg for people not at BP goal, and later in the program,for people of substantial body size, or with heavy resistantalbuminuria. Calcium channel blockers and diuretics were addedif needed for BP control and fluid retention. In the first yearwe used only low dose perindopril for people with serious renalinsufficiency, but thereafter abandoned the practice, findinglittle justification (10). The program also included healtheducation, and attempts to control blood glucose levels andlipids, when needed. After monthly visits for the initiationand titration of medicines, stable participants were seen every6 mo for the first 2 yr, and then once a year. Unstable patients,those not at BP goal or those with renal insufficiency wereseen as often as needed if they complied, sometimes every 2wk. BP were measured with cuffs appropriate for arm size, anda Dynamap machine with an electronic readout in the main clinic,and standard stethoscopes in the two smaller clinics.
After an initial training period in the clinic, local healthworkers and community liaison workers ran the program. Theyused community lists and recall systems, and algorithms fortesting and treatment; they also became familiar with the limitednumber of medicines used. They were supported, mostly remotely,by a nurse coordinator, with a doctor endorsing diagnoses andauthorizing treatment at a distance. The atmosphere of the programwas friendly rather than authoritative, and much activity happenedin the community as well as the clinic.
Terminal outcomes from natural causes of the treatment groupwere compared with those of historical controls who were followedin the pretreatment-program era, and whose baseline exam metone or more of the following criteria:
BP 140/90and diabetes or pathologic albuminuria(3.4+).
Diabetics withpathologic albuminuria (ACR 3.4+).
Overt albuminuria (ACR34+) regardless of BP or diabetes.
People with a single observation of elevated BP, but withoutother risk factors, were not included as controls.
Participants in the treatment program were followed until theydied or started dialysis, or through June 30, 2000, 4.56 yrafter the first person was enrolled. The course of controlswas followed against a background of no treatment or changingtreatment, from their baseline exam until they died, starteddialysis, or began systematic perindopril treatment for at least1 mo, or until June 30, 2000. Survivors among the controls werecensored 4.56 yr after their qualifying observation, the maximumfollow-up of people in the treatment program.
The fate of all persons was ascertained, and a cause of deathallocated by review of clinic and hospital records and deathcertificates. Only natural deaths were included as relevantterminal endpoints. These were categorized as all-cause naturaldeath (nonrenal and renal deaths), renal deaths (dialysis ordeath with chronic renal failure), nonrenal deaths, cardiovasculardeaths (primary or contributing), and nonrenal, noncardiovasculardeaths; these categories are described in more detail elsewhere(4). Unnatural deaths, due to accidents, suicide, homicide,drownings, and crocodile attack, were not included in analyzedendpoints: participants experiencing these endpoints were censoredfrom their cohorts at the time of the event.
Stata Statistical Software was used for data analysis (12).Terminal events were expressed as rates per 100 person-years.Hazard ratios for terminal events were calculated by the Coxproportional hazard method after checking that the assumptionsof that model were not violated. Survivals were also comparedby Kaplan Meier methods.
The treatment group consisted of 267 people, or about 30% ofall adults (20 yr). They were followed for a totalof 897 yr, with a mean (SD) of 3.36 (1.19) yr, with 233, 215,178, and 137 people reaching 6 mo, and 1, 2, and 3 yr of treatment,respectively. Dropouts included 23 who had died or developedrenal failure, 1 who died a suicide death, 7 who became normotensive,2 "false starts," 7 who became pregnant, 1 who was breast feeding,6 who had adverse effects (3 with cough, 2 with angioedema,1 with itching), 6 who decided to quit, 3 who moved, and 2 whowent onto palliative care.
A total of 327 people were identified as controls. They werefollowed for a total of 1041 yr, with a mean of 3.18 (1.48)yr. Of these, 171 (52.3%) were ultimately enrolled in the treatmentprogram. The other 156 were not enrolled in the treatment programbecause they refused treatment, eligibility criteria were notconfirmed, exclusion criteria were present, they died, or developeddeath or renal failure in the interim, or moved.
Table 1 shows that the treatment and control groups were fairlycomparable in terms of demographic and clinical status. Thetreatment group was marginally older than controls and was moreoften diabetic. Sixty two of the treatment group, or 23.2%,had been prescribed enalapril or captopril before enrollment.
Table 1. Characteristics of control and treatment groupsa
Of people on treatment at 3 yr, 2% were prescribed 2 mg perindopril,32% were prescribed 4 mg perindopril, and 66% were prescribed8 mg perindopril; 22.7% were also taking calcium channel blockersand/or diuretics. By self report, 66% were taking their medicines"most of the time"; 27%, "sometimes" or "occasionally"; and7%, "rarely or never" (13).
When the entire treatment group was followed over 3 yr, meanSBP and DBP fell within the first 6 mo, mean ACR did not changesignificantly, mean GFR appeared to rise, and mean serum potassiumincreased slightly. Table 2 shows the same trends in peoplewho were followed for the full 3 yr and had observations atevery interval. There was a swift sustained reduction in SBPand DBP. Mean ACR values did not change substantially. Serumcreatinine tended to rise over the first 1 or 2 yr, but subsidedto baseline level or lower at 3 yr, whereas calculated GFR showedreciprocal changes. Use of lab-adjusted values for creatinineaccentuated the apparent fall in creatinine and increase inGFR at 3 yr. Mean serum potassium rose slightly over the first2 yr, but no one developed hyperkalemia independent of sporadicsevere hyperglycemia, and there was never a need to modify perindoprildosage based on potassium levels.
Table 2. Clinical parameters in people passing 3 yr of treatment, unadjusted and avalues adjusted for different labs
Table 3 demonstrates treatment benefit at 3 yr in every categoryof participant. BP fell significantly in all groups except thosewith baseline GFR <60 ml/min. SBP fell by an average of 13.2mmHg in the aggregate group, and DBP fell by 6.8 mmHg. In peoplewho were hypertensive at baseline, SBP fell by a mean of 23mmHg, and DBP by a mean of 12.8 mmHg. There were significantfalls in BP in people who were "normotensive" at baseline, andin people who had previously been on ACEI, as well as in thosewho had not. ACR did not change significantly over 3 yr in anygroup, when the unadjusted values (shown) or the lab-adjustedvalues were considered. Unadjusted serum creatinine levels at3 yr tended to be unchanged or lower than at baseline, withthe exception of people with GFR <60 ml/min. Conversely,GFR tended to be stable or higher at 3 yr. Although these changesin unadjusted serum creatinine or GFR were NS, the fall in lab-adjustedserum creatinine levels were significant in all groups exceptdiabetics, people with low GFR, and those who had previouslybeen on ACEI treatment.
Table 3. Blood pressure and renal function (unadjusted) at baseline and 3 yr, by clinical features
Sixty one people reached a terminal event of natural cause:38 in the control group, and 23 in the treatment group. Twentyfive were renal deaths and 36 were nonrenal. Twenty eight deathshad a primary or underlying cardiovascular cause.
Figure 1 shows that rates of all-cause natural deaths were stronglyrelated to baseline ACR category, and were lower in the treatmentgroup than in controls for persons in every category of overtalbuminuria. No difference was apparent among the low deathrates of people with lower ACR.
Figure 1. Rates of all-cause natural death, by baseline albumin creatinine ratio (ACR) category, controls versus treatment cohort.
Table 4 shows an estimated 50% reduction in rates of all-causenatural deaths, and a 57% reduction in renal deaths in the treatmentgroup compared with controls, both of which are significant.It also shows a 46% reduction in nonrenal deaths, a 49% reductionin deaths with a cardiovascular component, and a 61% reductionin nonrenal, noncardiovascular deaths, although the associationswere NS at the P < 0.05 level.
Table 4. Terminal events (number/yr, and rate/100 yr (95% CI)) and hazard ratios (HR) by category of endpoint; treatment versus controls, adjusted for age, gender and baseline ACR
Although smaller numbers reduce the significance of the associations,and as shown in Table 5, there was apparent survival benefitin the treatment group in people with elevated BP and "normal"BP at baseline, in nondiabetics and diabetics, in people withall categories of overt albuminuria, and in those with wellpreserved renal function. Treatment benefit could not be discernedin people without overt albuminuria at baseline, and was marginalor nonexistent in people with GFR <60 ml/min.
Table 5. Numbers of all-cause natural death by patient characteristics and renal function, and hazard ratio, controls versus treatment, adjusted for age, gender, and ACR
Figure 2 compares the survival curves of treatment and controlgroups in people with overt albuminuria (in whom all of thesurvival benefit was segregated). The benefit of treatment tookabout 2 yr to appear. The estimated survival rate of the treatmentgroup at 4 yr was 86.6% versus 78.5% for the controls. Furtheranalyses showed a 4-yr survival for people with baseline ACRof 34 to 99 of 92% for the treated group and 87% for the controls;for people with ACR 100 to 199, survival rates were 85.4% and61.3% respectively, and for those with ACR 200+, the survivalrates were 75.7% and 45.9%, respectively.
Figure 2. Kaplan Meier survival curves, all-cause natural deaths, people with overt albuminuria (ACR, 34+), treatment versus controls.
Figure 3 shows that numbers of natural deaths and of the peoplestarting dialysis in the entire community began to fall soonafter the treatment program began. Aggregate data from otherAboriginal communities across the Top End of the NT, where screeningand treatment policies are changing more slowly, have not reflectedthe same sharp changes.
Systematic treatment of high-risk individuals in this communityresulted in marked improvement of BP and stabilization or improvementin renal function on a group basis. BP was very sensitive totreatment and the only people with hypertension that did notimprove were those where compliance was poor. This sequenceis markedly different from the deterioration of these parametersin the preprogram natural history.
Clinical parameters improved or stabilized in all groups ofpeople on treatment. People with all degrees of overt albuminuria,with normal and high BP, with and without diabetes, and withand without a previous history of ACEI treatment all showedbenefit, the last demonstrating the superior effect of a systematictreatment approach. Survival benefit was not evident among thevery low death rates of people with lower levels of urinaryalbumin, and was NS among the high death rates of people withlow GFR. However, the improvement in BP and reduction in progressionof ACR and loss of GFR in these groups suggest that survivalbenefit will become evident over a longer period of follow-up(14).
Treatment resulted in an estimated 50% reduction in rate ofnatural deaths and 57% reduction in rate of renal failure atan average follow-up of 3.39 yr. The use of historical controlsto estimate treatment effect on terminal events is imperfectfor several reasons (8), but trends in community-based deathsand renal failure supported the marked effect of the program.About 2 yr of treatment was required before reduction in all-causenatural death became apparent. This reflects, in part, the prioritizationof the sickest people for early entry into the program. It alsoimplies that the reductions in BP, progression of ACR, and lossof GFR that occur with treatment take time to translate intomeasurable survival benefits.
Impressive as they are, these figures understate the program’sfull potential benefit. Contributing to this understatementwas the prioritization of people with renal insufficiency whowere failing status quo (other ACEI-based) treatment for earlyenrollment in the program, and the use of low dose perindoprilin their management. The "intention-to-treat" analysis understatesfull potential treatment benefit, because only two-thirds weretaking their medicines most of the time. More rigorous medicalmanagement could improve outcomes even more: many people werenot at BP goal, control of blood glucose and lipid levels wasfar from optimal, angiotensin receptor blockers could be morevigorously added to ACEI therapy (15), and indications for treatmentof diabetics can be expanded (16).
Fewer renal deaths in the treatment group means lower morbidity,mortality, and costs in a community where 25% of all deathsare renal, and where most people with chronic renal failurego onto dialysis. However, the 46% reduction in nonrenal deathshas a tremendous effect in terms of premature death avoidedor postponed. Programs introduced in high-risk populations whereno prior treatment exists can be expected to have even greaterbenefit than ours, which was substituted for a previous lesssystematic treatment approach.
The predictive value of albuminuria for renal and nonrenal death(4) anticipates that rates should fall with an interventionthat retards albuminuria progression. A fall in cardiovasculardeaths is expected in view of the anti-hypertensive and cardiacand vascular-sparing effects of ACEI. However, the associationsof noncardiovascular nonrenal deaths with albuminuria and theapparent benefit with treatment are less readily explained.
A terminal endpoint was avoided in an estimated 23 people inthe treatment cohort—13 renal deaths and 10 nonrenal deaths.The number of people-needed-to-treat (NNT) over an average of3.4 yr to avoid a terminal endpoint was only 11.6: the NNT toavoid one renal death was 20.5 and the NNT to avoid one nonrenaldeath was 26.7. Thus the program was both effective and efficientin avoiding unwanted outcomes. It also saved costs. Baker etal. estimated the average annual cost per patient in the treatmentcohort over the first, most intensive, 3 yr of the program was$1,383Aus (17). Using data on dialysis costs and survival ondialysis (3,2) it was estimated that between $884,400 and $4,057,200was saved in those first 3 yr, through avoidance or postponementof dialysis alone. The range depends on what course ESRD incidencerates would have taken without intervention. Analysis throughmid-2000 suggests that the program continues to significantlyreduce dialysis costs. Reduced hospitalizations and a monetaryvalue placed on years of life gained in young and middle-agedadults would add to these savings.
Much of this program’s success derives from a strong senseof community ownership and control, a nonjudgemental, nonauthoritarianstyle, and respect for competing personal and community perspectivesand priorities. Individuals appreciate personalization of theirhealth goals, and many are slowly adopting lifestyle changes(18).
Most Aboriginal health services in remote Australia are under-resourced.Deficiencies in the prevention, surveillance, and managementof chronic diseases are especially notorious. This project hasshown that Aboriginal people are interested in health issues,willing to take medicine over the long term against a futurehealth risk, and can have an excellent clinical response. Italso shows that, in the specific instance of renal disease,considerable costs are saved by a systematic approach to regulartesting and treatment. These principles should be incorporatedinto regular adult health care in all Aboriginal communities(19) and resourced appropriately. Similar programs, varyingin details according to local realities, could also bring muchbenefit to high-risk populations in developing countries (20).
Acknowledgments
Research for this project was supported by Servier Australia,the Australian Kidney Foundation, Rio Tinto, Australia, theNHMRC of Australia, the Stanley Tipiloura Fund, Territory HealthServices, and the Colonial Foundation of Australia.
We are grateful for the support, enthusiasm, and participationof the Tiwi community, and the help of the staff of the TiwiIsland clinics at Nguiu, Milikapati, and Pirlangimpi. We alsothank the Tiwi Health Board for their review of this manuscript.We thank Susan Jacups and Kiernan McKendry for their dedicatedwork as treatment program coordinators, Jerome Kerinauia, andNellie Punguatji as Aboriginal Health workers; and Eric Tipilouraand Elizabeth Tipiloura as Community Project Officers. Finallywe thank Dr. Chris Harrison, the resident medical officer, forhis support and participation.
Footnotes
The participants in this program gave informed consent to havetheir course followed in the context of the projects "The epidemiologyand prevention of Aboriginal renal disease, Parts 1 and 2."These were approved by the Joint Institutional Ethics Committeeof the Menzies School of Health Research and Territory HealthServices, and its Aboriginal subcommittee, by the Tiwi LandCouncil (Part 1), and by the Tiwi Health Board (Part 2).
Cunningham J, Condon J: Premature mortality in Aboriginal adults in the Northern Territory. Med J Aust 165: 309–312, 1996[Medline]
Spencer JS, Silva D, Hoy WE: An epidemic of renal failure among Australian Aborigines. Med J Aust 168: 537–541, 1998[Medline]
You J, Hoy W, Beaver C, Zhao Y: Costs of hemodialysis and hospitalisations for patients with end stage renal disease in the Top End of the Northern Territory. Med J Aust 176: 461–465, 2002[Medline]
Hoy WE, Wang Z, Baker PRA, McDonald S, van Buynder PB, Mathews JD: The natural history of renal disease in Australian Aborigines. Part 2: The predictive value of albuminuria for premature death and renal failure. Kidney Int 60: 249–256, 2001[CrossRef][Medline]
Hoy WE, Pugsley DJ, Normal RJ, Hayhurst BG: A brief heath profile of adults in a Northern Territory Aboriginal community: with an emphasis on preventable morbidities. Aust NZ J Public Health 21: 121–126, 1997[Medline]
Hoy WE, Mathews JD, Pugsley DJ, McCredie DA, Hayhurst BG, Rees M, Walker KA, Kile E, Wang Z: The multidimensional nature of renal disease: Rates and associations of albuminuria in an Australian Aboriginal community. Kidney Int 54: 1296–1304, 1998[CrossRef][Medline]
Hoy WE, Baker P, Kelly A, Wang Z: Reducing premature death and renal failure in Australian Aborigines: Results of a community-based treatment program. Med J Aust 172: 473–478, 2000[Medline]
Hoy WE, Wang Z, Kelly A, Baker P RA: Sustained reduction in renal failure and cardiovascular deaths from a systematic treatment program in an Australian Aboriginal community. Kidney Int 63: [Suppl 83]: S66–S73, 2003
Cockcroft D, Gault MK: Prediction of creatinine clearance from serum creatinine. Nephron 16: 31–41, 1976[Medline]
The Sixth Report of the Joint National Committee on Prevention, Detection, Evaluation and Treatment of High Blood Pressure (JNC VI). Arch Intern Med 157: 2413–2446, 1997[Abstract]
Ruggenenti P, Perna A, Remuzzi G: ACE inhibitors to prevent end stage renal disease: When to start and why possibly never to stop: A post hoc analysis of the REIN trial results. J Am Soc Nephrol 12: 2832–2837, 2001[Abstract/Free Full Text]
Statacorp: Stata Statistical Software, Release 6.0. College Station, TX, Stata Corporation, 1999
Hoy WE, Baker P, Kelly A, Jacups S, McKendry K, McDonald S, Kerinauia J, Tipiloura E, Tipiloura E, Punguatji N, Harrison C, Wang Z: Stemming the tide: Reducing cardio-vascular disease and renal failure in Australian Aborigines. Aust NZ J Med 29: 480–483, 1999[Medline]
Hoy WE, Wang Z, Baker PRA, McDonald S, van Buynder PB, Mathews JD: The natural history of renal disease in Australian Aborigines. Part 1: Progression of albuminuria and loss of glomerular filtration rate over time. Kidney Int 60: 243–248, 2001[CrossRef][Medline]
Mackenzie HS, Ziai F, Omer SA, Nadim MK, Taal MW: Angiotensin receptor blockers in chronic renal disease: The promise of a bright future. J Amer Soc Nephrol 10 [Suppl 12]: S283–S286, 1999
The Heart Outcomes Prevention Evaluation Study Investigators: Effects of an angiotensin converting enzyme inhibitor, Ramipril, on death from cardiovascular causes, myocardial infarction and stroke in high-risk patients. N Engl J Med 342: 145–153, 2000[Abstract/Free Full Text]
Baker PRA: Cost-effectiveness analysis of a treatment program to prevent end stage renal disease in an Australian Aboriginal community. PhD thesis submitted through the University of Queensland, Australia, and the Menzies School of Health Research, Northern Territory, Australia, 2002
Kelly A, Tipiloura E, Hoy WE, et al: Compliance with a renal disease prevention program in an Australian Aboriginal community. Presented at the 35th MDSU Annual Scientific Meeting of the Australian New Zealand Society of Nephrology, 1999
Hoy WE: Screening and treatment for renal disease: The community model. Nephrology 4 [Suppl iii–iv]: S90–S95, 1998
Hoy WE: Reflections on the 15th International Congress of Nephrology. Renal and cardio-vascular protection in the developing world. Nephrol Dial Transplant 16: 1509–1511, 2001[Free Full Text]
Top
Abstract
Introduction
140/90 on two occasions), all diabetics with confirmed pathologic albuminuria (ACR, 
