Journal Logo

Original Articles: Clinical Transplantation

Cardiovascular Disease and Hypertension Risk in Living Kidney Donors: An Analysis of Health Administrative Data in Ontario, Canada

Garg, Amit X.1,2,3,11; Prasad, G V. Ramesh4; Thiessen-Philbrook, Heather R.1; Ping, Li3; Melo, Magda3; Gibney, Eric M.5; Knoll, Greg6; Karpinski, Martin7; Parikh, Chirag R.8; Gill, John9; Storsley, Leroy7; Vlasschaert, Meghan1,2; Mamdani, Muhammad3,10for the Donor Nephrectomy Outcomes Research (DONOR) Network

Author Information
doi: 10.1097/TP.0b013e31817ba9e3


As we promote living kidney donor transplantation as the preferred treatment option for end-stage renal failure, there is now global consensus that we also need better estimates of any long-term donor risks (1, 2). Early in the history of this procedure it was necessary and appropriate to quickly obtain data to ensure donation posed no great harm. However, existing studies cite variable estimates of morbidity (3, 4) and none have met epidemiologic standards for accurate risk assessment.

In an often-cited Swedish study, living kidney donors lived longer than the general population (5). However, donors go through rigorous evaluation to confirm good health, and risks would be better estimated by comparing a group of donors to the healthiest segment of the general population. Surprisingly, most of the existing studies did not use any type of control group to assess outcomes attributable to donation. To our knowledge, no published study has considered whether cardiovascular events are increased after kidney donation.

There are important reasons to evaluate the risk of cardiovascular disease. First, donors seem to have 5 mm Hg increase in blood pressure above that attributable to normal aging (3); in the general population every 10 mm Hg increase in systolic blood pressure or 5 mm Hg increase in diastolic blood pressure is associated with a one and a half-fold increase in mortality from both ischemic heart disease and stroke (6). This risk extends well into the “normal” range of blood pressure. Hypertension is one of the most important causes of cardiovascular disease worldwide, and was noted to be higher in one donor study (7) but not others (8–12). Second, in the general population, a glomerular filtration rate of 60 to 89 mL/min has been independently associated with premature cardiovascular disease (13). This new baseline level of glomerular filtration rate is realized by more than half of all kidney donors (4, 14). Finally, homocyst(e)ine and uric acid seem to increase after donation (15–17), and some consider these important factors in the development of atherosclerosis.

Using health administrative data, we observed a cohort to assess the risk of death and major cardiovascular events in living kidney donors. We also considered whether donors were more likely than controls to be diagnosed with hypertension during follow-up.


Design and Setting

We observed the cohort retrospectively by linking multiple administrative healthcare databases in the province of Ontario. We matched individual donors to randomly selected healthy residents who acted as controls (18). We aimed to select five controls for every donor. Ontario is Canada’s most populous province (38% of the Canadian population), with 10 to 12 million residents during the study period. Citizens of Ontario have universal access to hospital care and physician services through a single provincial government payer. In addition, anonymised healthcare records can be analyzed using encrypted identifiers to track individuals overtime. The study was conducted according to a prespecified protocol, and ethics approval was obtained from the institutional review board at Sunnybrook Health Sciences Centre, Toronto, Canada.

Data Sources

We used five databases: the Trillium Gift of Life Network Database recorded information for donors and recipients in Ontario undergoing kidney transplantation; the Canadian Institute for Health Information Discharge Abstract Database recorded hospital admissions including diagnostic and procedural information; the Ontario Health Insurance Plan Database provided information on physician and allied health claims for inpatient and outpatient services; the Registered Persons Database contained vital statistics on residents; and the Same Day Surgery Database recorded information on procedures such as fracture treatment not requiring overnight hospitalization. The latter four databases have been used extensively to research health outcomes (19–22). The datasets were reviewed from July 1, 1991 to March 31, 2006.

Donors and Controls

A total of 2033 individuals had a donor nephrectomy between July 1, 1993 and March 31, 2005. Some donors came from outside the province, and 1278 permanent residents could be linked to administrative data through their health card number and predonation information. There was no difference in the distribution of race or age at the time of donation, between those donors who could and could not be linked to provincial administrative data. Compared with those who could not be linked, linked donors were less likely to be male (40% vs. 48%).

Recipient relationship was recorded for all 1278 donors; 38% had donated to a sibling, whereas 18%, 15%, 12%, 9%, and 8% had donated to a parent, spouse, child, friend, or another relative, respectively. Racial information was available for about half of donors; 92% were White, 5% were Asian or Asian-Indian and less than 3% were African Canadian, Hispanic or Aboriginal Canadian. During the study period, individuals with pre-existing medical conditions such as hypertension or persistent hematuria were not readily accepted as donors in Ontario. Nondonor controls were randomly selected from healthy citizens residing in Ontario. Demographics were assessed at the time of donation. Baseline healthcare conditions and its use were assessed in the 2 years preceding donation. We matched donors and controls on age, sex, and neighborhood income (23). Income was categorized into quintiles, using the average income for each individual’s neighborhood at the time of transplantation. We also matched donors and controls on the number of non physician visits (optometrist, dentist, physiotherapy, chiropractor, or podiatry visits) to account for the propensity to seek and receive healthcare. We did not match on the number of physician visits, as donors have these visits for reasons of predonation assessment. To be eligible for selection, controls could not have any of the following: (a) A health condition such as hypertension, diabetes, cardiovascular disease, renal disease, or previous nephrectomy, which contraindicated donation; (b) An overnight hospitalization for any reason; and (c) More than 10 primary care visits. Because of a physician shortage, some individuals in Ontario did not have a primary care physician during the study period. To select controls with access to a physician for healthcare needs or routine assessment, all controls had at least two primary care visits in the 2 years preceding donation.

Outcomes: Death or Cardiovascular Disease, Hypertension

The primary outcome was a composite of time to death or major first cardiovascular event (myocardial infarction, stroke, coronary angioplasty, coronary bypass surgery, carotid endarterectomy, abdominal aortic aneurysm repair, or peripheral vascular bypass surgery) (Appendix A). The secondary outcome was time to a diagnosis of hypertension during follow-up (Appendix B). These outcomes were defined using codes proven to have good validity when health administrative data is compared with chart review (Appendix A, B).

Additional Analyses

We conducted several additional analyses for this study: (1) We separately assessed each component within the composite outcome of death and cardiovascular disease; (2) We assessed the outcome of hypertension using two other sets of validated diagnostic codes (24); (3) We considered if the observed association between donation and outcomes was different in four prespecified subgroups, stratifying the sample by age (<40 vs. >40 years of age), sex, socioeconomic status (lower vs. higher income), and whether the donation occurred before or after January 1, 2000; (4) We determined if the main outcomes differed between genetically related (i.e., siblings, parents, and children) and unrelated (i.e., spouses, friends) donors; (5) We checked for an association between donation and a few outcomes available within administrative data which are not biologically linked to nephrectomy but could be linked to health status and access to care. Specifically these outcomes were long-bone fracture, dermatology consultation, outpatient cholesterol testing, and immunization (predominantly influenza vaccination) (Appendix C). In addition, we quantified the number of primary care physician and nonphysician healthcare visits occurring after donation. These analyses were conducted as donors may see their physicians in follow-up more often than controls, increasing the likelihood that asymptomatic health conditions such as hypertension were diagnosed. We also conducted an unmatched analysis, which compared those donors and controls who had an average of two or more annual primary care physician visits during follow-up, and conducted a matched analysis adjusted for the average number of annual primary care physician visits during follow-up. Of note, donors in Ontario were not typically followed by their transplant professionals after the first year during the study period; (6) We assessed three other outcomes; time to first hospitalization for any reason, diagnosed diabetes mellitus, and new onset kidney failure (Appendix C); and (7) We repeated the analysis three times, varying the criteria used to randomly select healthy controls from citizens of Ontario, allowing the possibility of individuals with less primary care visits during the baseline period to be selected.

Statistical Analysis

To compare donors and controls on baseline characteristics, as described in Table 1 standardized differences were used to reflect the mean difference as a percentage of the standard deviation (25). For the purposes of statistical testing, a composite of death and major cardiovascular events was prespecified as the primary outcome, and hypertension as the secondary outcome. All other outcomes were specified in supplementary analyses. When six or fewer individuals developed an outcome of interest, exact numbers were not reported for reasons of privacy. Cox proportional hazard models were used to compute the association between donation and outcome after verification of the proportional hazards assumption. Matching variables were accounted for by the study design and the correlation of matched pairs were accounted for in the models; no additional variables were considered in the primary models. A supplementary analysis adjusted for the annual number of primary care visits in follow-up. Cox proportional hazard models were used to compare outcomes in genetically related and unrelated donors adjusting for gender, age, income based socio-economic status and the number of nonphysician visits in the 2 years before donation. A similar model was used to compare donors with those controls who had a median of two or more annual primary care physician visits during follow-up. In subgroup analyses, separate Cox proportional hazard models with interaction terms were used to determine whether the effect of donor status (donor vs. control) on the main outcomes was modified by age, sex, socioeconomic status, and year of donation; such analyses only had adequate statistical power for large effects and should be considered exploratory (26). Rates per 1000 patient years were compared using generalized estimating equations to account for the matched pairs (27). Sample size calculation suggested that 1200 donors matched to five controls per donor were required to identify a hazard ratio (HR) of 2.0 or more in death or cardiovascular events, based on a rate of 1.5% in controls (two tail alpha=0.05, beta=0.20) (28). Similarly, the same sample would identify a HR of 1.35 or more in diagnosed hypertension, based on a rate of 12% in controls (two tail alpha=0.01, beta 0.20). Statistical analyses were performed using SAS version 9.13 (SAS Institute Inc., Cary, NC).

Characteristics of donors and controls at the time of donation


Baseline Characteristics

We identified five suitable controls for more than 99% of the 1278 donors; a total of 6369 controls. Donors were similar to controls in most characteristics; the mean age was 41% and 60% were women. As expected, donors had more primary care physician visits (median [interquartile range] seven (4–11) vs. four (3–6) and cholesterol measurements [46% vs. 35%]) in the 2 years preceding donation (Table 1). The sample provided a total of 42,280 person-years of follow-up (mean 6.2 years, median 6.0 years, SD 3.2, range, 1-13). There was no difference in the mean follow-up time between donors and controls (6.2 vs. 6.2 years, P=0.95).

Primary Outcome: Death or Major Cardiovascular Event

There was no significant difference in death or cardiovascular events between donors and controls (1.3% vs. 1.7%; 2.0 vs. 2.7 events per 1000 person years; HR 0.7, 95% confidence interval [CI] 0.4-1.2, P=0.225) (Table 2).

Death or major cardiovascular events and hypertension among donors and controls

Secondary Outcome: Hypertension

Donors were more frequently diagnosed with hypertension than controls (16.3% vs. 11.9%; 29.1 vs. 20.6 events per 1000 person years; HR 1.4, 95% CI 1.2-1.7, P<0.001) (Table 2).

Additional Analyses

(1) There were no differences between donors and controls in any of the components of the composite outcome of death or cardiovascular disease (Table 2); (2) An increased risk of hypertension after donation was consistently observed using other sets of validated diagnostic codes (HR 1.4, 95% CI 1.2-1.7, and HR 1.6, 95% CI 1.4-1.8); (3) Donors were more frequently diagnosed with hypertension than controls in all subgroups, with no effect modification by age, sex, socioeconomic status, or year of donation (interaction term P values ranged, from 0.2 to 0.8); (4) There was no significant difference between genetically related (i.e., siblings, parents, children) and unrelated (i.e., spouses, friends) donors in death or cardiovascular events (1.2% vs. ≤1.6%; HR 0.9; 95% CI 0.5-3.0) or the diagnosis of hypertension (15.9% vs. 17.3%; HR 1.0; 95%CI 0.7, 1.3); (5) When compared with controls, donors were more frequently seen by their primary care physicians in follow-up (median number of annual visits 3.6, SD [4.7] vs. 2.6, SD [2.8]; 3563 vs. 2647 visits per 1000 person years; P<0.0001), but not by non physician healthcare providers (mean number of annual visits 1.1, SD [2.4] vs. 1.1 [2.7]; 1144 vs. 1132 visits per 1000 person years; P=0.91). There was no difference in time to fracture (3.1% vs. 2.6%; 5.1 vs. 4.3 events per 1000 person years; HR 1.2; 95% CI 0.8-1.7) or time to first dermatology consultation (15.6% vs. 14.4%; 36 vs. 30 events per 1000 person years; HR 1.1; 95% CI 0.9-1.3). However, donors were more frequently immunized compared with controls (291 vs. 188 immunizations per 1000 person years; P<0.0001) and received more outpatient cholesterol measurements (409 vs. 318 tests per 1000 person years, P<0.0001). An increased risk of hypertension after donation was evident when the analysis was restricted to only those donors and controls who had an average of two or more annual primary care physician visits during follow-up (HR 1.2; 95% CI 1.03-1.4). A borderline significant difference in hypertension diagnoses after donation was observed when the average number of annual primary care physician visits that occurred during follow-up was considered in the model (HR 1.1; 95%CI 1.0-1.3); (6) There was no difference in time to first hospitalization in donors compared with controls (17.1% vs. 17.0%; 30.9 vs. 30.7 events per 1000 person years; HR 1.0; 95% CI 0.9 1.2), or in a diagnosis of diabetes mellitus (2.7% vs. 2.5%; HR 1.1; 95% CI 0.8-1.6). Less than six individuals in each group developed end-stage renal failure; (7) When the analysis was repeated varying the criteria used to randomly select healthy controls from citizens of Ontario, the results were not appreciably different.


We observed a cohort using health administrative data to assess possible long-term medical risks to the living kidney donor. Our study provides important safety reassurances about the primary outcome, the potential risk of death and major cardiovascular events. This risk was unchanged in the first decade after kidney donation.

Our study extends current understanding by using a large sample of donors. To our knowledge, it is the first study to assess major cardiovascular events using validated codes within administrative data (Appendix A). Follow-up through universal health records is almost perfect; in other living donor studies on average 30% of donors were lost to follow-up (3). We matched individuals on important confounders, and this method of selecting healthy controls was better than most previous studies.

Yet, in the absence of randomization the appropriateness of the control group can still be questioned. Donors go through a detailed selection process and may be inherently healthier than even well selected controls. Conversely, certain outcomes may be ascertained more readily in donors than controls. We conducted additional analyses to explore these issues, and considered outcomes not biologically linked to nephrectomy but potentially linked to health status and access to care.

Limitations of Ontario’s health administrative databases are recognized. Many of the datasets are only reliable after the year 1990, restricting the observation period of the current analysis to about a decade. This analysis should be repeated in subsequent decades, to confirm cardiac events potentially attributable to nephrectomy are not taking longer to manifest. At this time it is reassuring that the results are similar to the published experience on 430 living donors followed through a registry in Sweden (5). In the Swedish study, deaths observed over 3 decades were no greater than expected for the general population. Unlike the Swedish study, the longevity of our donors was not statistically greater than controls; possibly because of the shorter follow-up time or because we used stricter criteria to select controls who represented the healthiest segment of the general population.

Family history was unavailable within the administrative databases, and we could not match donors to controls on their genetic susceptibility for illness. However, we did compare unrelated donors to those genetically related to the recipient, and there was no significant difference in outcomes. The data should not be generalized to non White donors, or to those with predonation health conditions such as hypertension which were contraindications to nephrectomy during the study period. Donors who could be linked to provincial administrative data differed in some demographics from those who could not. It is unlikely this resulted in important bias, as linkage was based on data collected at the time of nephrectomy and not on information collected during the follow-up period.

The secondary outcome of the study was a diagnosis of hypertension, as it was recorded in routine care. Hypertension was more frequently diagnosed in donors when compared with controls. A nephrectomy may predispose to hyperfiltration of the remaining kidney, and alterations in vascular tone, response to vasoactive peptides, sodium retention, and systemic regulation of angiotensin II (29, 30). All of these mediators may potentially lead to hypertension. Still, using claims data it is theoretically possible there were differences in thresholds to diagnosis or code hypertension. More relevant, more diagnoses may have been observed simply because blood pressure was monitored more carefully in donors when compared with controls. As demonstrated, donors were seen more frequently by their physicians, and were more likely to receive preventative healthcare such as immunizations and cholesterol measurements. Nevertheless, even if the more frequent diagnosis of hypertension was because of heightened surveillance among donors in follow-up, it carries additional considerations of medication prescriptions and costs, implications for insurability, and possible psychological stress (31).

Transplant professionals are currently conflicted about hypertension risk in donors; half believe the risk to be increased although the other half do not (32). In this analysis of health administrative data, rates of new-onset hypertension diagnoses were about one and a half-fold greater in donors when compared with controls. This can be compared with the results of six other controlled studies (7–12). Each of these studies enrolled between 15 to 63 donors and 15 to 50 controls. The control groups, some of whom had a genetic family history of kidney disease, were assembled at the time of donor follow-up evaluation. With the exception of one study (9), no studies seemed to follow control participants from the time of donor surgery. The results varied, the risk of hypertension was 2-fold higher in one study (7), but not statistically different in others (8–12). Yet, a decision to become a donor comes out of an intense desire to help a recipient, and most would disregard any warnings of an increased risk of hypertension (33). Most adults invariably develop hypertension in later life regardless if they were a donor. Nonetheless, based on existing data, it remains prudent to discuss a possible risk of hypertension which manifests earlier than it would otherwise. For those, select donors who do carefully consider risk-benefit or those circumstances where the recipient has strong preferences, disclosure of current study results might influence the decision to donate (33). For those, who consider accepting kidneys from altruistic strangers or paid donors, risk-benefit can also be considered (34). Given the possible risk, it remains prudent to counsel and regularly follow all donors to manage modifiable factors in an attempt to prevent hypertension and future cardiovascular disease. In this study, many, but not all donors visited their primary care physician on an annual basis. About 3% of donors were diagnosed with diabetes mellitus in the follow-up period.

In conclusion, in this study living kidney donors did not have a higher risk of death or major cardiovascular events when compared with appropriate controls in the first decade after kidney donation. The results from record linkage studies such as this will be best confirmed or refuted by resource intensive, prospective cohort studies. In such nonrandomized trials, large numbers of donors and controls will need to be recruited from multiple-centers, and followed for many years with detailed histories, examinations, and blood and urine testing. The ideal controls will be those individuals identified as potential donors who do not undergo the operation. Inclusion of racially diverse and older donors, including those with varying genetic susceptibility to kidney failure or with preexisting findings such as persistent hematuria, will help define whether there are differential effects of donation among these individuals. Such efforts will improve donor selection and the informed consent process, and guide-care which maintain good long-term health for previous and future living donors. Until such studies are completed, our results provide important safety assurances for those who consider living kidney donation, and for health professionals who care for such individuals.


All authors contributed to the design, analysis and interpretation of the study. Amit Garg, Heather-Thiessen Philbrook, Ping Li and Magda Melo had full access to all the data. Amit Garg had final responsibility for the decision to submit for publication, and is the guarantor. The opinions, results, and conclusions reported in this paper are those of the authors, and are independent from the funding sources. No endorsement from Ontario’s Ministry of Health and Long-Term Care, the Institute for Clinical Evaluative Sciences or Ontario’s Trillium Gift of Life is intended or should be inferred.


Definition of death, cardiovascular events using valid diagnostic and procedural codes for Canadian administrative data
Definition of hypertension diagnosis using valid diagnostic codes for Ontario administrative data
Definition of other outcomes


1. Ommen ES, Winston JA, Murphy B. Medical risks in living kidney donors: Absence of proof is not proof of absence. Clin J Am Soc Nephrol 2006; 1: 885.
2. Vastag B. Living-donor transplants reexamined: Experts cite growing concerns about safety of donors. JAMA 2003; 290: 181.
3. Boudville N, Prasad GV, Knoll G, et al. Meta-analysis: Risk for hypertension in living kidney donors. Ann Intern Med 2006; 145: 185.
4. Garg AX, Muirhead N, Knoll G, et al. Proteinuria and reduced kidney function in living kidney donors: A systematic review, meta-analysis, and meta-regression. Kidney Int 2006; 70: 1801.
5. Fehrman-Ekholm I, Elinder C-G, Stenbeck M, et al. Kidney donors live longer. Transplantation 1997; 64: 976.
6. Lewington S, Clarke R, Qizilbash N, et al. Age-specific relevance of usual blood pressure to vascular mortality: A meta-analysis of individual data for one million adults in 61 prospective studies. Lancet 2002; 360: 1903.
7. Watnick TJ, Jenkins RR, Rackoff P, et al. Microalbuminuria and hypertension in long-term renal donors. Transplantation 1988; 45: 59.
8. D’Almeida P, Keitel E, Bittar A, et al. Long-term evaluation of kidney donors. Transplant Proc 1996; 28: 93.
9. Miller IJ, Suthanthiran M, Riggio RR. Impact of renal donation. Long-term clinical and biochemical follow-up of living donors in a single center. Am J Med 1985; 79: 201.
10. Najarian JS, Chavers BM, McHugh LE, et al. 20 years or more of follow-up of living kidney donors. Lancet 1992; 340: 807.
11. Sobh M, Nabeeh A, el Din AS, et al. Long-term follow-up of the remaining kidney in living related kidney donors. Int Urol Nephrol 1989; 21: 547.
12. Williams SL, Oler J, Jorkasky DK. Long-term renal function in kidney donors: A comparison of donors and their siblings. Ann Intern Med 1986; 105: 1.
13. Tonelli M, Wiebe N, Culleton B, et al. Chronic kidney disease and mortality risk: A systematic review. J Am Soc Nephrol 2006; 17: 2034.
14. Ibrahim HN, Rogers T, Tello A, et al. The performance of three serum creatinine-based formulas in estimating GFR in former kidney donors. Am J Transplant 2006; 6: 1479.
15. Hida M, Iida T, Shimbo T. Renal function after nephrectomy in renal donors. Tokai J Exp Clin Med 1982; 7: 511.
16. Tsai MY, Aras O, Sozen H, et al. Plasma homocysteine levels in living kidney donors before and after uninephrectomy. J Lab Clin Med 2004; 143: 340.
17. Undurraga A, Roessler E, Arcos O, et al. Long-term follow-up of renal donors. Transplant Proc 1998; 30: 2283.
18. Costanza MC. Matching. Prev Med 1995; 24: 425.
19. Juurlink DN, Mamdani M, Kopp A, et al. Drug-drug interactions among elderly patients hospitalized for drug toxicity. JAMA 2003; 289: 1652.
20. Alter DA, Naylor CD, Austin P, et al. Effects of socioeconomic status on access to invasive cardiac procedures and on mortality after acute myocardial infarction. N Engl J Med 1999; 341: 1359.
21. Juurlink DN, Mamdani MM, Lee DS et al. Rates of hyperkalemia after publication of the Randomized Aldactone Evaluation Study. N Engl J Med 2004; 351: 543.
22. Mamdani M, Juurlink DN, Lee DS, et al. Cyclo-oxygenase-2 inhibitors versus non-selective non-steroidal anti-inflammatory drugs and congestive heart failure outcomes in elderly patients: A population-based cohort study. Lancet 2004; 363: 1751.
23. Bergstralh EJ, Kosanke JL. Technical Report Series No. 55, Computerized matching of cases to controls. Department of Health Science Research, Mayo Clinic, Rochester, Minnesota, 1995.
24. Tu K, Campbell NRC, Chen Z-L, et al. Validation of administrative databases in identifying patients with hypertension. Open Med 2007; 1: E3.
25. Mamdani M, Sykora K, Li P, et al. Reader’s guide to critical appraisal of cohort studies: 2. Assessing potential for confounding. BMJ 2005; 330: 960.
26. Pocock SJ. Clinical trials: A practical approach. Chichester, England, Wiley, 1983.
27. Liang KY, Zeger SL. Longitudinal data analysis using general linear models. Biometrika 1986; 73: 13.
28. Schoenfeld DA. Sample-size formula for the proportional-hazards regression model. Biometrics 1983; 39: 499.
29. Stein JH. Regulation of the renal circulation. Kidney Int 1990; 38: 571.
30. Hostetter TH, Olson JL, Rennke HG, et al. Hyperfiltration in remnant nephrons: A potentially adverse response to renal ablation. J Am Soc Nephrol 2001; 12: 1315.
31. Yang RC, Thiessen-Philbrook H, Klarenbach S, et al. Insurability of living organ donors: A systematic review. Am J Transplant 2007; 7: 1542.
32. Housawi AA, Young A, Boudville N, et al. Transplant professionals vary in the long-term medical risks they communicate to potential living kidney donors: An international survey. Nephrol Dial Transplant 2007.
33. Young A, Karpinski M, Treleaven D, et al. Tolerance for health risks to the living donor differs between potential donors, recipients and transplant professionals. Kidney Int 2008; 73: 1159.
34. Spital A. Public attitudes toward kidney donation by friends and altruistic strangers in the United States. Transplantation 2001; 71: 1061.
35. Cohen J. Statistical power analysis for the behavioural sciences. Hillsdale, New Jersey, Academic Press, 1988.
    36. Jha P, Deboer D, Sykora K, et al. Characteristics and mortality outcomes of thrombolysis trial participants and nonparticipants: A population-based comparison. J Am Coll Cardiol 1996; 27: 1335.
    37. Austin PC, Daly PA, Tu JV. A multicenter study of the coding accuracy of hospital discharge administrative data for patients admitted to cardiac care units in Ontario. Am Heart J 2002; 144: 290.
    38. Juurlink DN, Preya C, Croxford R, et al. Canadian Institute for Health Information Discharge Abstract Database: A validation study. Institute for Clinical Evaluative Sciences, Toronto, 2006.
      39. Liu L, Reeder B, Shuaib A, et al. Validity of stroke diagnosis on hospital discharge records in Saskatchewan, Canada: implications for stroke surveillance. Cerebrovasc Dis 1999; 9: 224.
      40. Richards J, Brown A, Homan C. The data quality study of the Canadian discharge abstract database. Proceedings of Statistics Canada Symposium 2001; Achieving Data Quality in a Statistical Agency, 2001.
        41. Canadian Institute for Health Information. Discharge abstract database data quality re-abstraction study: Combined findings for 1999/2000 and 2000/2001. Canadian Institute for Health Information, Ottawa 2002.
          42. Williams J, Young W. A summary of studies on the quality of health care administrative databases in Canada. In: Goel V, Williams JI, Anderson GM, et al, eds. Patterns of Health Care in Ontario. The ICES Practice Atlas. Ottawa, Canadian Medical Association, 1996.
            43. Raina P, Torrance-Rynard V, Wong M, et al. Agreement between self-reported and routinely collected health-care utilization data among seniors. Health Serv Res 2002; 37: 751.
            44. Potter BK, Manuel D, Speechley KN, et al. Is there value in using physician billing claims along with other administrative health care data to document the burden of adolescent injury? An exploratory investigation with comparison to self-reports in Ontario, Canada. BMC Health Serv Res 2005; 5: 15.
            45. Hemmelgarn BR, McAllister FA, Myers MG, et al. The 2005 Canadian Hypertension Education Program recommendations for the management of hypertension: Part 1. Blood pressure measurement, diagnosis and assessment of risk. Can J Cardiol 2005; 21: 645.
            46. Hux JE, Ivis F, Flintoft V, et al. Diabetes in Ontario: Determination of prevalence and incidence using a validated administrative data algorithm. Diabetes Care 2002; 25: 512.

            Cohort study; Living kidney donation; Hypertension; Cardiovascular disease

            © 2008 Lippincott Williams & Wilkins, Inc.