Screening for chronic kidney disease in Australia: a pilot study in the community and workplace

http://www.kidney-international.org

original article

& 2010 International Society of Nephrology

Screening for chronic kidney disease in Australia: a pilot study in the community and workplace Timothy H. Mathew1, Olivia Corso1, Marie Ludlow1, Adam Boyle1, Alan Cass2,3, Steven J. Chadban3,4, Beres Joyner5, Mark Shephard6 and Tim Usherwood3 1

Kidney Health Australia, Adelaide, South Australia, Australia; 2George Institute for International Health, Sydney, Australia; 3University of Sydney, New South Wales, Sydney, Australia; 4Department of Renal Medicine, Royal Prince Alfred Hospital, Sydney, Australia; 5 Department of Rehabilitation and Aged Care, Central Queensland Health District, Queensland, Australia and 6Community Point-of-Care Services, Flinders University Rural Clinical School, South Australia, Australia

The pilot program Kidney Evaluation for You (KEY) was conducted in Australia to screen for chronic kidney disease (CKD). Targeting people at high risk (those with diabetes, hypertension, a first-degree relative with kidney failure, or age 450 years), KEY aimed to establish community-based screening protocols, assess efficacy in promoting changes in risk-factor management, and explore participant CKD awareness. KEY offered free cardiovascular and kidney checks using point-of-care testing for on-site pathology measurements (estimated glomerular filtration rate, hemoglobin A1c, cholesterol, hemoglobin, albuminuria), lifestyle assessment, and exit interviews. Participants were telephoned at 3 months to ascertain whether KEY advice had been followed. Community and health professional support was strong; 99% of participants rated involvement as beneficial. Of 402 high-risk individuals recruited, findings were suggestive of CKD in 20.4%. Of these, 69% had hypertension, 30% diabetes, and 40% elevated total cholesterol. All participants with CKD stage 3b or higher were aged 461 years. Overall, 58% of participants were referred to their primary care providers for further action; of these, 82% saw their doctors in the next 3 months and 94% discussed KEY results. Follow-up telephone contact was successful for 82% of participants. A change in management occurred for 67%. Thus, the KEY approach to early detection of CKD and selected referral of participants was largely successful. Kidney International (2010) 77 (Suppl 116), S9–S16; doi:10.1038/ki.2009.538 KEYWORDS: Australia; chronic kidney disease; diabetes; hypertension; Kidney Early Evaluation Program; screening

Correspondence: Timothy H. Mathew, 25 North Tce, Hackney, South Australia 5069, Australia. E-mail: [email protected] Kidney International (2010) 77 (Suppl 116), S9–S16

The burden of chronic kidney disease (CKD) in Australia seems to be increasing. The rate of acceptance to Australian dialysis and transplant treatment programs has increased by about 140% in the last decade, and prevalence of patients on dialysis has more than doubled in the last 12 years.1 Prevalence of CKD stages in adult Australians (aged more than 25 years), as assessed in the AusDiab study, has been shown to be equivalent to prevalence in other developed countries; about 13.4% have CKD stages 1–5 and 7.8% have CKD stage 3–5.2,3 A recent overview of CKD in Australia includes information on prevalence, incidence, risk factors, health service use for CKD, CKD mortality, and the impact of CKD.3 Although the AusDiab survey has not been replicated, Australian prevalence of early CKD seems to have risen in recent years, driven by higher numbers of people with diabetes and the increasing age of the population. The situation is considered similar to that in the United States, where prevalence of CKD stages 1–4 increased from 10% in 1988–1994 to 13.1% in 1999–2004.4 Early recognition of CKD is made difficult by its largely asymptomatic nature. As one consequence of this, about a quarter of all patients in Australia present to their nephrologists with kidney failure less than 90 days before starting dialysis, and this fraction is not decreasing.5 Recent data in the United States Renal Data System (USRDS) 2009 Annual Data Report showed that only 57.2% of patients had visited nephrologists before initiation of renal replacement therapies.6 An early CKD diagnosis potentially allows exercise of preventive measures that may favorably affect clinical outcomes, and, for patients with progressive kidney failure, facilitate smooth, orderly, and less-morbid transition to dialysis or transplant. Evidence supports a strategy for control of CKD through early detection and intensive management of three risk factors for CKD and its progression: diabetes, high blood pressure, and protein in the urine. An Australian study showed that in primary care, opportunistic screening of patients aged 50–69 years and intensive management of screen-detected patients with diabetes, high blood pressure, or protein in the urine is likely to be a cost-effective strategy S9

original article

for preventing kidney failure and cardiovascular morbidity and mortality.7 Systematic screening for CKD is not widely practiced in Australia. Screening is accomplished through simple tests, such as a urine test for albuminuria or proteinuria, a blood test for creatinine, and blood pressure measurement. It is generally agreed that screening the whole population for CKD is not cost-effective, and that any screening approach should target a high-risk segment of the population.8 Targeted screening can occur in two ways: community- or workplace-based screening with testing offered to all who choose to enrol, and opportunistic screening in primary-care settings with testing offered to people without CKD symptoms who present to the health-care system for other reasons. The relative merits of these two approaches in Australia have recently been reviewed.9 The World Health Organization principles of screening for disease were published 40 years ago,10 and continue to underpin current screening recommendations in Australia.11 Screening for CKD is considered to fulfill most of these criteria. In June 2007, Kidney Health Australia received funding from BHP Billiton and the Commonwealth Department of Health and Ageing to conduct a pilot study of a targeted CKD early detection program in the Australian community and workplace. The principal objectives of the Kidney Evaluation for You (KEY) project were the following: K To test an effective and affordable means of finding early CKD in high-risk individuals and referring them to primary health-care providers for appropriate long-term care. K To promote preventative health behaviors by increasing awareness of modifiable risk factors for CKD (and other chronic diseases with common risk factors) in the sampled communities. K To engage primary health-care providers in the early detection and management of CKD. K To raise awareness of CKD in the tested populations. The project aimed to provide free kidney health checks in up to four selected communities (at least one each metropolitan, regional, and remote) over a 1-week period to produce a final sample size of 400 participants. A national Steering Committee was formed to oversee development of KEY in relation to methodology, educational materials, project evaluation, and preparation of the final report, within agreed-upon timeframes and in collaboration with Kidney Health Australia staff nominated to assist with this project. In the United States, after an initial pilot program that screened 889 individuals at increased risk for kidney disease in 21 cities from April through to October 1997, the US National Kidney Foundation (NKF) launched a nation-wide community-based kidney health screening program (the Kidney Early Evaluation Program, KEEP) in August 2000.12 At present in its ninth year of operation, NKF KEEP has screened more than 128,000 participants. KEEP targets S10

TH Mathew et al.: Targeted CKD screening in Australia

populations aged 18 years and older with a history of diabetes or hypertension or a first-order relative with diabetes, hypertension, or kidney disease. Because of comparable eligibility criteria, comparing participant characteristics and biochemical test results between KEY and KEEP under a common set of disease definitions and health screening questions is research relevant, even though KEY has just begun with a pilot study. RESULTS

Of 421 participants initially enrolled, 402 were eligible for inclusion in the study and were assigned to mutually exclusive eligibility categories according to the following hierarchy: aboriginal or Torres Strait Islander, 2%; previous diagnosis of diabetes, 18%; previous diagnosis of hypertension, 35%; age 450 years, 40%; first-degree relative with end-stage kidney disease, 5%. About half of the participants had multiple risks for CKD, including 4% with three or more risk factors. Fifteen participants (4%) indicated previous kidney disease diagnoses. The mean age of participants was 58.0 (±11.1) years; 84% were aged 450 years. Of the sample, 53% were men, approximately half had completed at least 11 years of schooling, and 59% were employed. Most participants (85%) reported seeing a general practitioner regularly, and 90% of these had visited their doctors in the last 6 months; 87% of participants agreed to have KEY findings forwarded to their nominated general practitioners. Comparable descriptive data for NKF KEEP participants and additional demographic data for KEY participants, including a breakdown by location of testing, are found in Table 1. One-third of participants considered their health to be very good or excellent, compared with 29% in NKF KEEP. Of the sample, 12% were smokers and 50% consumed more alcohol than the recommended two standard drinks per day; 16% indicated that they ate the recommended five servings of vegetables per day, and 25% added salt to their cooking or to food at the table. Obesity was diagnosed in 44% of participants, and mean body mass index was 30.0 (±4.7) kg/m2 for men and 28.5 (±6.1) kg/m2 for women. Awareness of selected facts about CKD was low. Only 14% of participants knew that one ‘could lose up to 90% of kidney function before feeling sick,’ and 13% correctly identified that a third of adult Australians are at increased risk of CKD. However, common risk factors for developing CKD were well known, with 84–92% of participants identifying these correctly. Similarly, in NKF KEEP, 11.8% of participants were current smokers and 44.4% were obese (body mass index X30 kg/m2), with mean body mass index 29.5 (±5.7) kg/m2 for men and 30.5 (±7.3) kg/m2 for women. Proteinuria (one plus or more by dipstick) or microalbuminuria (by quantitative analysis) was found in 13% of participants, and hematuria (by dipstick) in 13% (19% of women, 8% of men). Urine albumin–creatinine ratio (ACR) was not determined in the presence of hematuria, dipstick proteinuria, or positive dipstick leukocyte/nitrite reading, on Kidney International (2010) 77 (Suppl 116), S9–S16

original article

TH Mathew et al.: Targeted CKD screening in Australia

Table 1 | Demographic characteristics of Kidney Evaluation for You (KEY) and National Kidney Foundation (NKF) Kidney Early Evaluation Program (KEEP) participants KEY sites Characteristics

Total (n=402)

Townsville (n=166)

Roxby Downs (n=143)

Perth (n=93)

Age (mean±s.d., range) 58.0±11.1, 20–89 62.7±9.4, 32–82 53.6±9.8, 23–81 56.6±12.5, 20–89 Men 53 (214) 44 69 47 Aboriginal and Torres Strait Islander 2 (8) 3 2 0 Australian or English ancestry 70 (283) 76 66 67 Elevated blood pressuref 43 (173) 67 64 42 15 (60) 20 8 14 Diabetesg 12 (49) — — — Current smokerh Dyslipidemia 34 (135) 39 32 33 Screening evidence suggestive of CKD 20 (82) 24 18 18 Completed year 11 schooling or equivalent 49 (196) 34 57 64 Employed 59 (236) 37 79 70 Retired 36 (144) 59 16 27 Have regular doctor 85 (334) 95 71 87 91 90 88 Saw doctor in last 6 months 90 (282/313)k Gave permission to send results to 87 (348) 94 83 79 nominated doctor

P-value 0.000b 0.000c NS NS

0.000i 0.000i 0.000b 0.000j NS 0.00b

NKF KEEPa n=89,622 54.1±15.5, 18–108 31.7 4.6d 49.9e 43.7 29.4 11.8 63.9 26.2 84.4 NA NA 86.7 71.6l 74.3

Abbreviations: CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; NA, not applicable; NS, nonsignificant. Note: Data are presented as % (n) or % unless otherwise noted. a 2000–2008. b Townsville4Roxby Downs and Perth. c Roxby Downs4Townsville and Perth. d Non-Hispanic Native American. e Non-Hispanic white. f Treated and not treated. g Known diabetes plus elevated A1c (KEY) or elevated blood glucose (KEEP). h KEY smoking information not available by site. i TownsvilleoRoxby Downs and Perth. j Roxby DownsoTownsville and Perth. k Data missing for nine participants. l Within the last year.

the basis of the manufacturer’s recommendation. Of the 300 participants with interpretable urine ACR, 8% had microalbuminuria. In contrast, microalbuminuria (ACR X30 mg/g) was found in 23.6% when the test was performed on all NKF KEEP participants. Findings suggestive of CKD were detected in 82 participants (20.4% overall, 23% of men, 17% of women). Of these, 48% were categorized as stage 3 (Figure 1a). Subdividing stage 3 into 3a (estimated glomerular filtration rate 45–59 ml/min per 1.73 m2) and 3b (estimated glomerular filtration rate 30–44 ml/min per 1.73 m2) showed 38 and 10% with results suggestive of each, respectively. The yield of CKD was age related; all participants with findings suggestive of stage 3b or higher were aged X61 years (Figure 2). NKF KEEP CKD prevalence data by stage are illustrated in Figure 1b. The yield of CKD was also age related; 58% of participants with CKD were aged 460 years, and 81% of this group had CKD stage 3 or higher. Chronic kidney disease presence was associated with hypertension (treated or untreated) in 69% of KEY and 76.4% of NKF KEEP participants, with diabetes (defined as being on medication for diabetes or with abnormal hemoglobin A1c (HbA1c)) in 30% of KEY participants, and with elevated total cholesterol or taking cholesterollowering medication in 40% of KEY and 59.5% of NKF KEEP Kidney International (2010) 77 (Suppl 116), S9–S16

participants. The HbA1c test was introduced in NKF KEEP only in June 2008; however, presence of CKD was associated with diabetes defined as on medication for diabetes or with an elevated glucose level (fasting blood sugar X126 mg/dl, nonfasting blood sugar X200 mg/dl) in 24.4% of NKF KEEP participants. In KEY, blood pressure greater than 140/90 mm Hg was detected in 43% of participants (vs 43.7% in NKF KEEP). An additional 17% (vs 24.3% in NKF KEEP) had normal blood pressure readings but reported being on blood pressure medication. Thus, 60% of KEY participants (vs 68% in NKF KEEP) were considered hypertensive; new hypertension was diagnosed for 24% (vs 15.3% in NFK KEEP). Of participants already taking antihypertensive medication, blood pressure readings were not at target levels for 51% (Figure 3). Rates of untreated hypertension were highest among participants aged 31–45 years (28%) and 46–60 years (30%). The location factor for untreated hypertension was striking; 46% of participants from a rural mining town had untreated elevated blood pressure, compared with 14% from a regional center and 17% from a city. Participants from the rural mining town were significantly younger than other participants, and were more likely to be men and less likely to report having a regular doctor. Hemoglobin A1c testing was performed for all 402 participants, and only one elevated HbA1c (47%) was found S11

original article

TH Mathew et al.: Targeted CKD screening in Australia

Percent of CKD population

among the 321 participants without previously diagnosed diabetes. Elevated HbA1c levels above treatment target were detected in 31 of 76 participants with known diabetes, 29 of whom were already on diabetes medication. Elevated HbA1c

50 45 40 35 30 25 20 15 10 5 0

38 29 22 10

1

1

0 5

2

3a

3b CKD stage

4

2

3a

3b CKD stage

4

60

Percent of CKD population

50

40

30

20

10

0 1

5

Figure 1 | Prevalence (%) of chronic kidney disease stages. Prevalence (%) of chronic kidney disease stages in KEY (a, n ¼ 82/402) and National Kidney Foundation Kidney Early Evaluation Program (b, n ¼ 20,200/77,080). CKD, chronic kidney disease.

levels above treatment target were more common in men (19%) than women (10%), and at ages 61–75 years (23%). Elevated total cholesterol (defined as a total cholesterol 45.5 mmol/l in people with diabetes and 46.5 mmol/l in people without diabetes) was found in 12% of KEY participants (vs 44.5% in NKF KEEP) overall. An additional 22% (vs 19.4% in NKF KEEP) had normal cholesterol readings but reported taking medication for hypercholesterolemia. Of the 94 KEY participants who reported taking medication to lower cholesterol, 6% (vs 30.6% in NKF KEEP) had a cholesterol reading above target. Of KEY participants not on cholesterol-lowering medication (n ¼ 307), only 41 (13 vs 47% in NKF KEEP) had elevated cholesterol. In KEY, 4% of participants (vs 11.6% in NKF KEEP) had hemoglobin levels diagnostic of anemia. For 9 of the 16 participants with anemia, findings suggestive of CKD were present, a yield of anemia in the CKD group of 11% (vs 18% in NKF KEEP). Follow-up by telephone at 3 months after assessment was successful for 82% of participants (vs 30% in NKF KEEP). The rate of successful follow-up was highest among participants from the regional center (90%), and lowest among those from the metropolitan location (70%). The rate was also low among younger participants (33% for ages 18–30 years compared with 86% for ages 46–60 years). Follow-up was accomplished equally when assessed by sex, employment status, and referral to a general practitioner. Of the 402 participants, 58% were referred to see their general practitioners, because of at least one abnormality in CKD or cardiovascular risk assessment. Of those who were referred, 82% (vs 71% in NKF KEEP) reported visiting their general practitioners in the next 3 months, and 94% (vs 67% in NKF KEEP) of these reported discussing screening results (Figure 4). Of participants not specifically referred back to

60

Percentage

50 40 30 20 10 0 18 – 30 years (n = 6)

31 – 45 years 46 – 60 years (n = 39) (n = 202)

61 – 75 years (n = 127)

> 75 years (n = 27)

Stage 1

50

10

3

3

4

Stage 2

0

10

5

6

11

Stage 3a

0

3

4

13

22

Stage 3b

0

0

0

6

0

Stage 4

0

0

0

0

4

Figure 2 | Prevalence (%) of chronic kidney disease stages for each age category in KEY (n ¼ 402). All participants at stage 3b or higher (estimated glomerular filtration rate 15–44 ml/min per 1.73 m2) were aged 61 years or older. S12

Kidney International (2010) 77 (Suppl 116), S9–S16

original article

TH Mathew et al.: Targeted CKD screening in Australia

n = 380 (data missing for 22 participants)

On antihypertensive medication n = 137 36% (KEEP 52.7%)

Normal blood pressure n = 67, 49% (KEEP 24.3%)

High blood pressure n = 70, 51% (KEEP 28.4%)

Not on antihypertensive medication n = 243 64% (KEEP 47.3%)

Normal blood pressure n = 151, 62% (KEEP 31.9%)

High blood pressure n = 92, 38% (KEEP 15.3%)

Figure 3 | Flow chart for self-reported antihypertension medication status and blood pressure readings in KEY and National Kidney Foundation Kidney Early Evaluation Program (KEEP).

All participants, n = 402 Follow-up successful, n = 329

Referred to general practitioner n = 188

Not referred to general practitioner n = 141

Visited general practitioner n = 155

Visited general practitioner n = 97

Discussed KEY results n = 146

Discussed KEY results n = 78

No changes initiated by general practioner n = 37

1+ change initiated by general practitioner n = 109

1+ change initiated by general practitioner n = 40

No changes initiated by general practioner n = 38

Figure 4 | Flow chart for physician visits and further tests conducted after KEY.

their general practitioners because of KEY-detected abnormalities, 69% visited their general practitioners in the next 3 months, and 80% of these reported discussing KEY results. Management changes (new or changed prescription, changes in lifestyle, early review) were self-reported by 67% of the 224 participants who visited their doctors and discussed the KEY results. A new medication was prescribed for 21% of this group, and 8% were referred to specialists. For 86% of participants who were referred back to their doctors and who discussed the KEY findings, at least one new test was ordered or initiated by the doctor at that visit (including blood and urine tests, cystoscopy, and kidney/ bladder ultrasounds). The 34 participants with findings suggestive of CKD stage 3 or higher are of particular interest; 30 of these went back to their doctors, 29 reported discussing the KEY findings, and 20 (59%) made management changes as a consequence. Self-reported lifestyle changes were initiated as a consequence of KEY participation by 46% of those with follow-up information. Changes included improved nutrition (e.g., responses indicating more healthful eating through reduced fat content, salt content, or portion sizes) reported by 33% of Kidney International (2010) 77 (Suppl 116), S9–S16

participants, increased exercise reported by 27%, weight loss reported by 19%, and smoking cessation reported by 3%. Awareness of CKD facts was reassessed at telephone followup; increased awareness was found, particularly regarding general kidney knowledge and CKD risk factors. Of KEY participants who completed follow-up, 99% (vs 99.3% in NKF KEEP) reported that taking part had been beneficial for them. The point-of-care pathology testing received favorable comment; 99% of participants reflected on its convenience and 96% reported that the immediate results assisted them in understanding their condition. DISCUSSION

The KEY program successfully recruited 402 eligible participants from three diverse Australian communities: a rural mining town, a regional center, and a metropolitan capital city. Recruitment targets were readily achieved at each site despite a limited, low-budget advertising campaign using local newspapers, radio stations, and posters. A condition of eligibility for KEY was the presence of at least one specified risk factor for CKD. About half of the participants had multiple risk factors, including 4% with 3 or S13

original article

more; 84% of participants were aged older than 50 years, and half of them had no other risk factor. Previously diagnosed hypertension was present in 47% of the KEY cohort, compared with 30% prevalence found in adult Australians in the AusDiab study; diabetes was present in 19% of the KEY cohort, compared with 7.6% in the Australian population.3 Our findings are similar to NKF KEEP and to other programs reported in this supplement, demonstrating widespread agreement among screening programs targeted at high-risk populations.13–15 Participant demographic profiles differed by site (Table 1). Regional center participants were older, less educated, less likely to be employed, and less likely to be men. These differences reflect the fact that regional center enrolment was entirely from the community, whereas the workplace contributed a significant proportion of recruits at the other two sites. The finding of evidence suggestive of CKD in 20.4% of the 402 eligible participants reflects the targeted nature of this study. The yield from the whole community in the AusDiab study (adults aged 425 years) was 13.4%.3 The KEY CKD yield was somewhat lower than that in NKF KEEP, which reported a yield of 27.1% in a cohort of 61,675 participants using similar eligibility criteria.16 Updated CKD prevalence reported in the 2008 NKF KEEP Annual Data Report was 26.2% among 89,622 participants.15 This may reflect the absence of African-American participants, who are overrepresented in KEEP, and among whom CKD prevalence is known to be high. The low recruitment rate of indigenous people (well recognized to have a heavy burden of CKD) in KEY reflected the choice of recruitment sites, an issue that should be addressed in future programs. CKD yield was highest in the regional center (24%), where participants were on an average 9 years older than participants in the rural mining town and 6 years older than those at the metropolitan site, findings in keeping with the well-recognized increasing CKD prevalence with age. The age cutoff for recruitment to KEY was set arbitrarily at 50 years. The yield of moderateto-severe CKD would have been substantially increased with a higher age cutoff, as all participants with findings suggestive of CKD stage 3b or higher were aged older than 61 years. Of participants aged less than 50 years, 15 of 16 with findings suggestive of CKD had stages 1 or 2. These observations suggest that targeted screening program criteria may serve better if based on a decision tree analysis in which age, diabetes, hypertension, and cardiovascular disease interact to determine the best yield, versus traditional risk-factor approaches. The USRDS recently reported a decision tree analysis suggesting age cutoff at X60 years, and that people aged less than 60 years with diabetes and hypertension should be targeted.17 Two methodological characteristics of the KEY protocol make our estimation of CKD prevalence in this sample a conservative one. We chose not to include hematuria in our definition of CKD for the purpose of this survey, although it is included in the accepted Kidney Disease Outcomes Quality Initiative (KDOQI) definition of CKD.18 In part, this was in S14

TH Mathew et al.: Targeted CKD screening in Australia

recognition of the fact that hematuria needs to be shown to be persistent (about half of repeat tests are negative), and to be shown to reflect kidney damage and not a urological cause, particularly in an elderly population. We also followed the manufacturer’s (Siemens, Munich, Germany) instruction that microalbuminuria testing should not be performed in the presence of significant proteinuria, hematuria, or leucocyturia. This decision excluded 25% of participants from microalbuminuria assessment. A unique component of the KEY study was the use of point-of-care testing for pathology measurements rather than transporting blood and urine samples to a centralized laboratory for analysis. This approach was chosen for several reasons, including the recent availability of a validated and IDMS-aligned creatinine measurement on a point-of-care testing device.19 One trained operator completed measurements on five point-of-care testing devices (separately measuring blood cholesterol, creatinine, HbA1c, hemoglobin, and urine ACR) within 10 min, and could handle specimens from up to 30 participants each day. Other attractive features of point-of-care testing include the capacity to inform participants about their results on the spot, and to formulate and discuss with them appropriate advice and recommendations. Training nurses to be competent in using point-of-care testing devices was readily accomplished in a 1-day training session. The overlap of cardiovascular disease, diabetes, and CKD is such that the KEY program was considered to be an appropriate opportunity to assess CVD risk in participants. However, each additional test adds cost, time, and complexity to the program. The yield of new diabetes was strikingly low, with only one new case found among 321 nondiabetic participants. The yield of abnormal cholesterol levels was also low. Similarly, anemia was present in only 16 of 401 participants, and 9 of them had CKD. All anemia cases were mild. Blood pressure was found to be above target levels in 42% of participants. Of those already taking antihypertensive medication, blood pressure was above target levels for 51%. Untreated hypertension was found in 42% of participants from the rural mining town, three times the proportion found in the regional center. The AusDiab study finding of untreated hypertension in the Australian adult population was 15%.20 KEY findings are consistent with NKF KEEP findings, in which 40% of the entire cohort and 54% of participants with prescribed antihypertension medication had elevated blood pressure. These findings indicate substantial potential for improvement in blood pressure management for participants, supporting the targeted nature of the program. The high rate of contact and discussion by KEY participants with their own doctors can be interpreted as evidence that the information the program provided was perceived as important and contributed to overall care. Most participants had recently seen their general practitioners. Thus, it would seem that we were relatively unsuccessful in Kidney International (2010) 77 (Suppl 116), S9–S16

TH Mathew et al.: Targeted CKD screening in Australia

reaching people with poor access to primary health care, a shortcoming also demonstrated in NKF KEEP. Both programs seemed to attract highly motivated individuals with good access to the health-care system. Change in medical management (new or changed prescription, changed lifestyle, early review) was reported by 67% of the 224 participants who visited their doctors and discussed KEY results, suggesting that KEY participation might have improved health outcomes. However, the full impact of the KEY program is difficult to determine, as these results are based on self-report data only. Longer-term follow-up of KEY participants would help determine whether changes lead to sustained improvement in risk-factor control and potential progression of kidney disease. Several limitations apply to interpreting and using data from the KEY project. These include small sample size, potential selection bias through recruitment by advertising, self-determination of eligibility, short follow-up time, and self-reporting of outcomes. These limitations were intrinsic to the pilot study. It was not designed or funded to provide evidence of improved physical outcomes. Such evidence would allow assessment of the cost-effectiveness of this targeted approach to the early detection of asymptomatic CKD. Owing to the limitations, any conclusions from this study must be interpreted with caution. This pilot study of a targeted community and workplace CKD screening program was highly successful in recruiting participants, engaging local health professionals in education sessions, generating significant publicity through local media despite a limited advertising budget, and detecting CKD. Connection back to local doctors was successful, and changed management as a consequence of KEY participation was reported by two-thirds of participants. However, given the substantial cost of mounting a program such as KEY, future studies would be needed to establish the cost-effectiveness of targeted screening and potential changes in age cutoff to improve the yield. The alternative way of early CKD detection through targeted opportunistic screening in general practice has already been established to be cost-effective7 and may prove to be the most sustainable approach in this country.9 The most important observations, however, relate to poor risk-factor control in the CKD population, supporting the need for primary care to address this high-risk population. Future studies from multiple countries will be needed to determine effectiveness and the ability to motivate the general public and physicians to assess CKD in targeted groups. MATERIALS AND METHODS The Ethics Committee of the Australian Government Department of Health and Ageing approved the study, and written informed consent was obtained from all participants before testing. Data collection, storage, and use were consistent with the National Privacy Principles in the Privacy Act 1988. Once informed consent was obtained, each participant was allocated a unique four-digit identification number. This was recorded on all data collection forms and was used to code the data. Kidney International (2010) 77 (Suppl 116), S9–S16

original article

During early 2008, the KEY program was offered to three geographically diverse communities: Townsville, Queensland, a regional center with a population of about 150,000 people; the Roxby Downs region, South Australia, a rural mining town with a population of about 4500 people; and Perth, West Australia, a metropolitan capital city with a population of about 1.5 million people. Participation in the project was promoted through printed materials and newspaper articles, and through radio and television interviews. Individuals aged 50 years or older (or 35 years or older if of aboriginal and/or Torres Strait Islander descent), with a firstdegree relative with known end-stage kidney disease, or known to have high blood pressure (treated or not) or diabetes were eligible to participate. Eligible individuals used a free-call 1800 phone number to book appointments. Participants from the rural mining town and the metropolitan city included a mix of mining company employees (the company was a joint sponsor of the project) and members of the general public. Before or during the KEY testing period, CKD education workshops were held for local health professionals to inform them of the KEY program and the importance of early CKD detection. The KEY testing procedure consisted of a series of work stations operated by a renal nurse or scientist, and were held at a community venue or company workplace. Participants moved through the testing stations, had blood and urine samples tested on site, and received all results by the end of the appointment. Blood pressure measurements were performed in accordance with clinical guidelines produced by the National Heart Foundation of Australia,21 with hypertension (more strictly an ‘elevated blood pressure,’ as this study included only one blood pressure reading) defined as systolic blood pressure X140 mm Hg, diastolic blood pressure X90 mm Hg, or current use of antihypertensive drugs irrespective of blood pressure reading. Point-of-care testing devices were used to analyze blood and urine (abnormal cutoff levels in parentheses) for the following clinical factors: K I-STAT (Abbott Point of Care, Winsor, NJ, USA) for blood creatinine, (4105 mmol/l, men; 485 mmol/l, women) and random glucose (47 mmol/l); K DCA 2000 þ (Siemens Healthcare Diagnostics, Deerfield, IL, USA) for HbA1c (47%) and urine ACR (42.5 mg/mmol, men; 43.5 mg/mmol, women); K Cholestech LDX System (Cholestech, Hayward, CA, USA) for total cholesterol (45.5 mmol/l with diabetes, 46.5 mmol/l without diabetes); ¨ ngelholm, Sweden) for K HemoCue Hb 201 þ (HemoCue AB, A hemoglobin (o130 g/l, men; o120 g/l, women). Urine dipstick analysis for glucose, ketones, blood, protein, nitrite, and leukocytes preceded urine ACR measurements. Daily analytical performance checks on all point-of-care testing devices using quality control materials were performed. The blood creatinine measurements obtained from the iSTAT device were aligned to the IDMS standard and bias was corrected after a detailed validation study was performed. Estimated GFR in ml/min per 1.73 m2 was calculated using the Modification of Diet in Renal Disease Study22 formula. CKD was defined and staged in accordance with the KDOQI guidelines.18 The single eGFR result obtained during KEY could not establish a CKD diagnosis; abnormal results were interpreted to participants and their local doctors as suggesting CKD and requiring confirmation. S15

original article

Additional clinical factors recorded included age, sex, body mass index, waist–hip ratio, reported diagnosis of hypertension and use of antihypertensive therapy, diabetes and use of antidiabetic medications, high cholesterol and use of lipid-lowering drugs, and family history of end-stage kidney disease. Recorded lifestyle factors included smoking behavior, alcohol consumption, physical exercise, salt habits, and daily fruit and vegetable intake. At an exit interview with a renal nurse, participants were advised to visit a general practitioner if their clinical results were categorized as abnormal. With participant permission, test results were forwarded to the nominated general practitioner. Each participant was provided a copy of the test results and a package of educational material. Three months after completing KEY, participants were followed up by telephone (with a minimum of three contact attempts) to obtain details regarding subsequent visits to health-care providers, reassess CKD awareness, and record any clinical interventions that had occurred as a result of KEY participation. Statistical analyses were performed using SPSS (version12.0; SPSS, Chicago, IL, USA). Differences between continuous variables were tested using one-way analysis of variance. Differences between categorical variables were tested using w2-test (when comparing two groups) or general linear models (when comparing three or more groups). McNemar’s test was used to assess differences between two paired groups (changes in kidney awareness at follow-up). The Bonferroni test was used for all post hoc comparisons. A two-tailed significance level of 0.05 was used for all tests.

TH Mathew et al.: Targeted CKD screening in Australia

3.

4. 5.

6.

7.

8. 9.

10. 11.

12.

13.

DISCLOSURE

Dr Cass has received consulting fees and grant support from Roche, and lecture fees from Servier. Dr Chadban has received consulting fees from Novartis, Wyeth, and Janssen-Cileg, lecture fees from Novartis, Roche, and Amgen, and grant support from Amgen. Dr Usherwood has served on the Board of Directors of WentWest Ltd, and has received grant support from WentWest Ltd, General Practice Ltd, Pfizer Australia, and Brain Resource Company Ltd. The other authors declare no competing interests. ACKNOWLEDGMENTS

Financial support for this study was provided by the Australian Government Department of Health and Ageing; BHP Billiton Pty Ltd. Alan Cass is supported by a National Health and Medical Research Council Senior Research Fellowship. We thank Allan J. Collins, MD, FACP, for perspective and insight, Shu-Cheng Chen, MS, of the Chronic Disease Research Group, for help with NKF KEEP data, and Shane Nygaard, BA, and Nan Booth, MSW, MPH, of the Chronic Disease Research Group, for manuscript preparation and manuscript editing, respectively. REFERENCES 1. McDonald S, Excell L, Livingston B. Australia & New Zealand Dialysis and Transplant Registry (ANZDATA) 2008 Annual Report. Chapter 1: Stock and Flow. Thirty-First Report. 2008. Adelaide, South Australia. 2. Chadban SJ, Briganti EM, Kerr PG et al. Prevalence of kidney damage in Australian adults: the AusDiab kidney study. J Am Soc Nephrol 2003; 14(Suppl 2): S131–S138.

S16

14.

15.

16.

17.

18.

19. 20.

21.

22.

Australian Institute of Health and Welfare. An overview of chronic kidney disease in Australia 2009. Available at: http://www.aihw.gov.au/ publications. Accessed 9 November 2009. Coresh J, Selvin E, Stevens LA et al. Prevalence of chronic kidney disease in the United States. JAMA 2007; 298: 2038–2047. McDonald S, Excell L, Livingston B. Australia & New Zealand Dialysis and Transplant Registry (ANZDATA) 2008 Annual Report. Chapter 2: New Patients. Thirty-First Report. 2008. Adelaide, South Australia. U.S. Renal Data System. USRDS 2009 Annual Data Report: Atlas of Chronic Kidney Disease & End-Stage Renal Disease in the United States, 2009 edn. National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases: Bethesda, MD, 2009. Howard K, Salkeld G, White S et al. Kidney Health Australia: the costeffectiveness of early detection and intervention to prevent the progression of CKD in Australia. Available at http://www.kidney.org.au/assets/documents/ Stage%202%20Costing%20Study%20CKD%20preventionFINAL.pdf. Accessed 9 November 2009. Boulware LE, Jaar BG, Tarver-Carr ME et al. Screening for proteinuria in US adults: a cost-effectiveness analysis. JAMA 2003; 290: 3101–3114. Mathew T, Corso O. Review article: early detection of chronic kidney disease in Australia: which way to go? Nephrology (Carlton) 2009; 14: 367–373. Wilson JMG, Junger G. Principles and practice of screening for disease. 34. 1968. World Health Organization. Australian Population Health Department Principal Committee (Screening Subcommittee). Population based screening framework. Australian Health Ministers’ Advisory Council 2008. Available at http:// www.cancerscreening.gov.au/internet/screening/publishing.nsf/Content/ pop-based-screening-fwork/$File/screening-framework.pdf. Accessed 9 Novwember 2009. Brown WW, Peters RM, Ohmit SE et al. Early detection of kidney disease in community settings: the Kidney Early Evaluation Program (KEEP). Am J Kidney Dis 2003; 42: 22–35. Takahashi S, Okada K, Yanai M. The Kidney Early Evaluation Program (KEEP) of Japan: results from the initial screening period. Kidney Int 2010; 77(Suppl 116): S17–S23. Obrador GT, Garcı´a-Garcı´a G, Villa AR et al. Prevalence of chronic kidney disease in the Kidney Early Evaluation Program (KEEP) Me´xico and comparison with KEEP US. Kidney Int 2010; 77(Suppl 116): S2–S8. National Kidney Foundation. Kidney Early Evaluation Program (KEEP) 2008 Annual Data Report. Am J Kidney Dis 2009; 53(Suppl 4): S1–S135. Whaley-Connell AT, Sowers JR, Stevens LA et al. CKD in the United States: Kidney Early Evaluation Program (KEEP) and National Health and Nutrition Examination Survey (NHANES) 1999–2004. Am J Kidney Dis 2008; 51(Suppl 2): S13–S20. U.S. Renal Data System. USRDS 2008 Annual Data Report: Atlas of Chronic Kidney Disease & End-Stage Renal Disease in the United States, vol. 1 National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases: Bethesda, MD, USA, 2008. National Kidney Foundation. K/DOQI clinical practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Kidney Disease Outcome Quality Initiative. Am J Kidney Dis 2002; 39(Suppl 2): S1–S246. Shephard M, Corso O, Mathew T. Creatinine can be measured at point-of-care. (abstract). Nephrology 2008; 13(Suppl 3): A101. Briganti EM, Shaw JE, Chadban SJ et al. Untreated hypertension among Australian adults: the 1999–2000 Australian Diabetes, Obesity and Lifestyle Study (AusDiab). Med J Aust 2003; 179: 135–139. Guide to management of hypertension 2008, Assessing and managing raised blood pressure in adults. Available at http://www. heartfoundation.org.au. Accessed 9 November 2009. Levey AS, Coresh J, Greene T et al. Using standardized serum creatinine values in the modification of diet in renal disease study equation for estimating glomerular filtration rate. Ann Intern Med 2006; 145: 247–254.

Kidney International (2010) 77 (Suppl 116), S9–S16

Copyright of Kidney International is the property of Nature Publishing Group and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.

Copyright of Kidney International Supplement is the property of Nature Publishing Group and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.