Original Report: Transplantation American
Journal of
Nephrology
Received: May 17, 2014 Accepted: June 7, 2014 Published online: September 2, 2014
Am J Nephrol 2014;40:174–183 DOI: 10.1159/000365157
Understanding Antihypertensive Medication Use after Living Kidney Donation through Linked National Registry and Pharmacy Claims Data Krista L. Lentine a, b Mark A. Schnitzler a, b Amit X. Garg f Huiling Xiao a David Axelrod d Janet E. Tuttle-Newhall b Daniel C. Brennan c Dorry L. Segev e
a
Center for Outcomes Research and b Division of Abdominal Transplantation, Department of Surgery, Saint Louis University School of Medicine, c Division of Nephrology, Department of Medicine, Washington University School of Medicine, St. Louis, Mo., d Division of Abdominal Transplantation, Department of Surgery, Dartmouth Hitchcock Medical Center, Hanover, N.H., and e Division of Abdominal Transplantation, Department of Surgery, Johns Hopkins University, Baltimore, Md., USA; f Division of Nephrology, Western University, London, Ont., Canada
Abstract Background: Use of antihypertensive medications (AHM) after living kidney donation is not well described. Methods: We examined a database wherein national transplant registry data for 4,650 living kidney donors in 1987–2007 were linked to pharmacy claims from a US private health insurer (2000–2007 claims) to identify post-donation AHM fills. Cox regression with left- and right-censoring was used to estimate the frequencies and relative likelihood (adjusted hazards ratios, aHR) of post-donation AHM fills according to donor demographic traits. Medication possession ratio (MPRs), defined as (days of AHM dispensed)/(days observed), were also compared among donors and non-donor general beneficiaries. Results: Overall, 17.8% of the sample filled at least one AHM by 5 years post-donation. As compared with White living donors, African-Americans had 37% higher relative likelihood of any AHM use after donation (aHR 1.37, p < 0.0007), including significantly higher likelihoods of filling
© 2014 S. Karger AG, Basel 0250–8095/14/0402–0174$39.50/0 E-Mail [email protected] www.karger.com/ajn
diuretics (aHR 2.25, p < 0.0001), ACEi/ARBs (aHR 1.46, p < 0.01), calcium channel blockers (aHR 1.56, p = 0.03), and vasodilators/other agents (aHR 2.17, p = 0.03). MPRs for any AHM and subcategories were lower among donors compared with age- and sex-matched non-donors. However, AHM MPRs rose in donors with multiple hypertension diagnoses, and prescription fill exposure for all AHM classes except diuretics was similar among donors and general nondonors with ≥3 hypertension diagnoses. Conclusions: While AHM requirements are lower after kidney donation than among unscreened general persons, racial variation in AHM use occurs in privately insured donors. Demonstration of pharmaceutical care needs of insured donors supports the need for long-term follow-up and healthcare access for all donors. © 2014 S. Karger AG, Basel
Introduction
In the context of the organ shortage, kidney transplantation from living donors has increased markedly over the last several decades [1] and has evolved to more commonly include donors with baseline medical complexity Krista L. Lentine, MD, PhD Saint Louis University, Salus Center 4th Floor 3545 Lafayette Avenue, St. Louis, MO 63104 (USA) E-Mail lentinek @ slu.edu
Downloaded by: Univ. of California Santa Barbara 198.143.33.33 - 8/29/2015 12:43:21 PM
Key Words Antihypertensive agents · Hypertension · Insurance · Kidney transplantation · Living donors · Pharmacy claims · Registries
and greater racial and ethnic diversity [2, 3]. Importantly, despite the growth and expansion of living donation, mandated follow-up of living donors in the USA has been limited in scope, duration and completeness. Currently, the Organ Procurement and Transplantation Network (OPTN) collects data on living donors for only 2 years of follow-up [4], and incomplete reporting and donor loss to follow-up have been common [5, 6]. While OPTN/ United Network for Organ Sharing (UNOS) policies were modified in 2013 to include mandated thresholds for collection of follow-up information by centers, compliance with these thresholds could still leave missing data (up to 20% missing for clinical and 30% for laboratory results) [4], and the predictive value of information at 2 years for important dimensions of long-term donor health is uncertain. Hypertension is an intermediate health outcome warranting better understanding after live kidney donation. Given established associations of hypertension as a leading cause of chronic kidney disease and end-stage renal disease (ESRD) in the general population [7], there is theoretical concern that post-donation hypertension may accelerate glomerular filtration rate decline in donors. Hypertension and renal function are reciprocally related, and data from predominantly White cohorts suggest increased hypertension risk in prior donors compared to the general population, possibly due to physiological alterations (hyperfiltration in the remaining kidney, changes in vascular tone and renin-angiotensin-aldosterone regulation) and/or heightened follow-up [8, 9]. Based on integration of the OPTN registry and administrative billing claims from public and private insurers, we recently found racial variation in post-donation hypertension diagnoses, with 40–50% higher relative risks of hypertension diagnoses in African-American versus White donors, including two to three times the relative risks of malignant hypertension [10, 11]. A recent report of 103 African-American donors at two centers suggested that the frequency of post-donation hypertension may exceed that of matched controls, noting a high proportion of previously undiagnosed hypertension identified through study encounters [12]. Further, a new study of data for 96,217 US living donors identified a small but significant increase in ESRD risk attributable to donation that was highest in African-Americans [13], illustrating the need to understand mediators of disparities in post-donation ESRD. Management of hypertension may include lifestyle modification and close observation, followed by pharmacologic therapy with antihypertensive medications (AHM) in patients who do not achieve treatment
goals [14, 15]. To date, limited information has been reported on AHM use in demographically diverse donor samples. Pharmacy claims offer a non-obtrusive measure of prescribed healthcare that do not rely on patient self-report and are increasingly used in observational investigations of large populations including transplant-related epidemiologic studies [16–21]. To advance understanding of medically treated hypertension after live kidney donation, we examined a linkage of OPTN registry data for living donors with administrative records from a private health insurer that includes pharmacy claims for prescription medications. Our primary aim was to identify post-donation AHM fills among a large sample of US living donors, overall and by agent class, and to compare post-donation AHM fills within donors according to race. In secondary analyses, we also compared AHM exposure among donors to that in age- and sex-matched general non-donors overall, as well as fill patterns among donors and non-donors with clinical hypertension diagnoses.
Antihypertensive Medication Use in Living Kidney Donors
Am J Nephrol 2014;40:174–183 DOI: 10.1159/000365157
Data Source and Sample This study used data from the OPTN. The OPTN data system includes information on all donors, waitlisted candidates, and transplant recipients in the USA, submitted by the members of the OPTN. The Health Resources and Services Administration (HRSA), US Department of Health and Human Services provides oversight to the activities of the OPTN contractor. Study data were assembled by linking OPTN records for living kidney donors with administrative data from a national private health insurer. After approval by HRSA and the Saint Louis University Institutional Review Board, beneficiary identifier numbers from the insurer’s electronic databases were linked using names and birthdates to unique OPTN identifiers for living kidney donors. Analyses were performed using Health Information Portability and Accountability Act-compliant, limited datasets with all direct identifiers removed. We included living kidney donors who had records of donating between October 1987 and July 2007 and benefits under the participating insurer after donor nephrectomy at some point in May 2000 to December 2007 (the period of available claims data). All study participants were simultaneously enrolled in medical and pharmacy benefits with this insurer exclusively during the study period. Because of the large sample size, the anonymity of the patients studied, and the non-intrusive nature of the research, a waiver of informed consent was granted per the Department of Health and Human Services Code of Federal Regulations (Title 45, Part 46, Paragraph 46.116). Non-donors were sampled from records of general insurance beneficiaries enrolled in the same plan at some point during May 2000 to December 2007.
175
Downloaded by: Univ. of California Santa Barbara 198.143.33.33 - 8/29/2015 12:43:21 PM
Methods
Statistical Analyses Correlates of AHM Use after Living Kidney Donation Datasets were merged and analyzed with SAS for Windows software, version 9.3 (SAS Institute, Inc., Cary, N.C., USA). As windows of captured insurance benefits varied across the sample, Cox regression with left- and right-censoring was used to estimate the frequency of patients with AMH fills over time after donation, and associations (adjusted hazards ratios, aHR) between donor traits (particularly race) and AHM fills. Censoring was applied from donation to insurance enrollment and after the end of an individual’s captured insurance benefits (or end of study, December 2007). Comparison of AHM MPRs in All Donors and General Non-Donors To compare AHM exposure after donation to use among a non-donor sample as a secondary analysis, living donors were matched one-to-one with general insurance beneficiaries by sex and age when benefits began. Maximum observation time, defined by benefits duration in each matched pair, was limited to the shortest available in the pair. We compared MPRs for any AHM and AHM classes in donors and matched general non-donors using the paired Wilcoxon signed-rank test. A schematic of the design for subject matching and AHM MPR computation among the full donor cohort and age- and sex-matched non-donors is shown in figure 1a. Comparison of AHM MPRs in Donors and General Non-Donors with Hypertension We also sought to describe AHM exposure among living donors and non-donors with clinical hypertension diagnoses. Hypertension diagnoses were ascertained from medical billing claims with corresponding International Classification of Disease, Ninth
176
Am J Nephrol 2014;40:174–183 DOI: 10.1159/000365157
Revision, Clinical Modification (ICD-9-CM) diagnosis codes and subclassified as benign (401.1, 402.10, 402.11, 403.10, 403.11, 404.10–404.13, 405.11, 405.19), malignant (401.0, 402.00, 402.01, 403.00, 403.01, 404.00–404.03, 405.01, 405.09) or unspecified (401.9, 402.90, 402.91, 403.90, 403.91, 404.90–404.93, 405.91, 405.99), as previously described [11]. Living donors with hypertension were matched to non-donors with hypertension by age at hypertension diagnosis, sex, and hypertension type (fig. 1b). Here observation time for MPR computation began at the first captured hypertension diagnosis. Finally, we examined AHM exposure patterns among living donors with multiple post-donation hypertension diagnoses as an index of severity by repeating matching procedures and AHM MPR computations among donors and nondonors with ≥2 and ≥3 captured hypertension diagnoses. Observation time for AHM MPR computation in these analyses began at the date of the second and third captured hypertension diagnosis, respectively.
Results
Demographic Correlates of AHM Use after Living Kidney Donation Characteristics of the 4,650 live kidney donors in the study cohort have been previously reported [10]. Among the sample, 13.1% were African-American, 76.3% White, 8.2% Hispanic, and 2.4% other races (table 1). Mean age at donation was 37.2 ± 10 years. The median times from donation to the start and end of observed insurance eligibility were 4.9 and 7.7 years, respectively. Distributions of race and sex in the linked donor sample were similar to that of all living kidney donors in the OPTN in the period [10]. Overall, at 5 years post-donation, 17.8% (95% CI 15.1– 20.0) of the sample filled prescriptions for at least one AHM, including fills for diuretics in 7.2% (95% CI 5.5– 8.9) and ACEi/ARBs in 5.9% (95% CI 4.4–7.4). By 5 years, 6.6% (95% CI 5.0–8.3) of the donor sample had received β-blockers, 3.4% (95% CI 2.2–4.6) had filled calcium channel blocker prescriptions, and 1.2% (95% CI 0.05– 1.9) had received vasodilators/other agents. In multivariate regression including adjustment for donor age and sex, African-American race was associated with 37% higher likelihood of fills for any AHM after donation (aHR 1.37, 95% CI 1.09–1.72) (table 1). Compared with White donors, African-American donors also had more than twice the likelihoods of filling diuretics (aHR 2.25, 95% CI 1.71–2.95) and vasodilators/other agents (aHR 2.17, 95% CI 1.07–4.42), as well as significantly higher likelihood of filling ACEi/ARBs (aHR 1.46, 95% CI 1.08–1.99) and calcium channel blockers (aHR 1.58, 95% CI 1.03–2.42) after donation. Lentine/Schnitzler/Garg/Xiao/Axelrod/ Tuttle-Newhall/Brennan/Segev
Downloaded by: Univ. of California Santa Barbara 198.143.33.33 - 8/29/2015 12:43:21 PM
Definitions of Outcomes and Covariates The primary outcome of interest was pharmacy claims for AHM fills after donation. AHM were also subcategorized as ACE inhibitors/angiotensin II receptor blockers (ACEi/ARBs), βblockers, calcium channel blockers, diuretics or vasodilators/other agents. Combination pills (e.g. diuretics with other AHM agents) were considered to contribute mediation to each of the component classes. Primary analyses considered time to first fill of any AHM or AHM class, respectively, among the donor sample. In secondary analyses, we computed medication possession ratios (MPRs), a metric quantifying the fraction of days of captured insurance enrollment for which AHM were prescribed [19– 21]. MPRs were defined as: [days of medication supplied over an observation window]/[days of observation], where observation windows were defined as the period of captured insurance benefits for an individual. To account for concomitant use of multiple agents, the MPR metric for any AHM exposure aggregated fill days from different classes even if prescription dates were overlapping, such that the maximum possible MPR for any AHM may exceed 1.0. Demographic data from the OPTN at donor nephrectomy included age, gender, and race as reported by the transplant center. The insurance records include information on age and sex, but not race; thus race information was not available for non-donors. Body mass index was reported to the OPTN for only 5.3% of the linked donors and so was inadequate for analysis.
Color version available online
Match by: Age at INS enrollment, sex Non-donor INS benefits AHM MPR
Days fill
Days fill
Days fill
Days fill
Days fill
Days of AHM Observation days
Days of AHM Observation days
LKD INS benefits
a
Start observation at INS enrollment (at or after donation)
Limit observation to minimum in matched pair
Match by: Age at HTN Dx, sex, HTN type Non-donor INS benefits AHM MPR
Days fill
Days fill
HTN Dx in medical claims Days fill
Fig. 1. Schematic of design for AHM medi-
Days fill
Days of AHM Observation days
LKD INS benefits
b
Start observation at HTN Dx
Limit observation to minimum time after HTN Dx in matched pair
AHM MPRs in All Donors and General Non-Donors AHM exposure in donors and age- and sex-matched general non-donors was compared using MPRs, the fraction of days of captured insurance enrollment for which AHM were prescribed. Overall, AHM MPRs were lower among the full donor sample than among age- and sexmatched non-donors: any AHM, 11 vs. 26%, p < 0.00001; diuretics, 3 vs. 8%, p < 0.0001; ACEi/ARBs, 4 vs. 11%, p < 0.0001; β-blockers, 3 vs. 6%; calcium channel blockers, 2 vs. 4%, and vasodilators/other agents, 1 vs. 0%. Stratify-
ing the full donor cohort by time from nephrectomy to insurance enrollment demonstrated higher MPRs among live donors enrolled after, compared with before, the median time to enrollment: any AHM, 18 vs. 5%; diuretics, 5 vs. 1%; ACEi/ARBs, 6 vs. 2%; β-blockers, 5 vs. 1%, calcium channel blockers, 2 vs. 1%, and vasodilators/other agents, 1 vs. 0% (fig. 2). However, MPRs remained lower in donors compared with their age- and sex-matched non-donor controls after stratification by enrollment timing.
Antihypertensive Medication Use in Living Kidney Donors
Am J Nephrol 2014;40:174–183 DOI: 10.1159/000365157
177
Downloaded by: Univ. of California Santa Barbara 198.143.33.33 - 8/29/2015 12:43:21 PM
cation possession ratio computation among (a) the full living donor cohort and ageand sex-matched general non-donors and (b) donors and non-donors with hypertension diagnoses, matched by age at diagnosis, sex and hypertension type. AHM = Antihypertensive medication; Dx = diagnosis; HTN = hypertension; INS = insurance plan; LKD = living kidney donor; MPR = medication possession ratio.
Days fill
Days of AHM Observation days
LKDs General controls
MPR (%)
40 30 20
*
10 0
*
* Any
Diuretics
ACEi/ARBs
*
*
BBs
CCBs
†† Others
LKDs enrolled after 4.9 years post-donation
50 40 MPR (%)
Fig. 2. AHM medication possession ratios
among living donors compared with ageand sex-matched general non-donor controls, stratified by time from donation to study enrollment as below or above the sample median. ACEi/ARBs = ACE inhibitors/angiotensin II receptor blockers; BBs = β-blockers; CCBs = calcium channel blockers; LKD = living kidney donor; MPR = medication possession ratio. † p = 0.05; ‡ p < 0.05–0.0001; * p < 0.0001.
Color version available online
LKDs enrolled within 4.9 years post-donation
50
30
*
20 10 0
*
* Any
Diuretics
* ACEi/ARBs
BBs
CCBs
† Others
Table 1. Adjusted associations of baseline demographic factors with likelihoods of post-donation pharmacy fills for any AMH and AHM
subclasses in privately insured living kidney donors Trait distributions Age at donation (per year) Male gender Race White, Non-Hispanic African-American Hispanic Other
any AHM
diuretics
ACEi/ARBs
CCBs
BBs
vasodilators/ others
37.2±10.0# 1.05 (1.04–1.06)§ 1.04 (1.03–1.05)§ 1.06 (1.05–1.07)§ 1.05 (1.04–1.07)§ 1.06 (1.04–1.07)§ 1.06 (1.03–1.09)§ 45.4% 0.95 (0.82–1.11) 0.63 (0.50–0.78)§ 1.30 (1.05–1.61)* 0.83 (0.61–1.13) 1.13 (0.87–1.46) 1.07 (0.61–1.87) 76.3% 13.1% 8.2% 81.2%
reference 1.37 (1.09–1.72)* 1.02 (0.74–1.40) 0.47 (0.21–1.05)
reference 2.25 (1.71–2.95)§ 1.07 (0.68–1.70) 0.37 (0.09–1.51)
reference 1.46 (1.08–1.99)* 1.18 (0.77–1.81) 0.52 (0.17–1.64)
reference 1.58 (1.03–2.42)* 0.82 (0.40–1.68) 0.72 (0.18–2.92)
reference 1.07 (0.71–1.61) 0.74 (0.40–1.37) 0.22 (0.03–1.57)
reference 2.17 (1.07–4.42)* 0.36 (0.052.66) 1.16 (0.16–8.49)
ACEi/ARBs = ACE inhibitors/angiotensin II receptor blockers; BBs = β-Blockers; CCBs = calcium channel blockers. # Mean ± SD; * p < 0.05–0.0001; ≤ 0.0001.
AHM MPRs in Donors and General Non-Donors with Hypertension MPRs computed among subgroups identified by the presence of clinical hypertension diagnoses were, as expected, substantially higher than among the full cohort overall. The MPR for any AHM exposure in donors with 178
Am J Nephrol 2014;40:174–183 DOI: 10.1159/000365157
at least one captured hypertension diagnosis was 61% (fig. 3). Further, there were graded increases in AHM MPRs among donors after increasing numbers of hypertension diagnoses, rising to 72% days exposed to any AHM in donors with at least two diagnoses and 90% in those with at least 3 diagnoses. MPRs for any AHM and Lentine/Schnitzler/Garg/Xiao/Axelrod/ Tuttle-Newhall/Brennan/Segev
Downloaded by: Univ. of California Santa Barbara 198.143.33.33 - 8/29/2015 12:43:21 PM
§ p
AHM pharmacy fill outcomes, aHR (95% CI)
0.9
Color version available online
Care after at least 1 HTN diagnosis
1.0
*
0.8
MPR
0.7 0.6 0.5 0.4
††
0.3
*
††
0.2 0.1 0
1.0 0.9
Any
Diuretics
ACEi/ARBs
BBs
CCBs
Others
Care after at least 2 HTN diagnoses
*
0.8 0.7 MPR
0.6 0.5
*
0.4
††
0.3 0.2
† †
††
BBs
CCBs
0.1
among living donors and general non-donors with clinical hypertension diagnoses, matched by age at hypertension diagnosis, sex and hypertension type (number per group by diagnosis frequency: ≥1, 894; ≥2, 672; ≥3, 500). ACEi/ARBs = ACE inhibitors/angiotensin II receptor blockers; BBs = β-blockers; CCBs = calcium channel blockers; HTN = hypertension; LKD = living kidney donor; MPR = medication possession ratio. ‡ p < 0.05–0.0001; * p < 0.0001.
MPR
Fig. 3. AHM medication possession ratios
1.1 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0
Any
Diuretics
ACEi/ARBs
Others
Care after at least 3 HTN diagnoses
*
LKDs General controls
††
Any
Diuretics
ACEi/ARBs
BBs
CCBs
Others
most classes remained lower in donors compared with non-donors matched for age at first hypertension diagnosis, sex and coded hypertension type, with the exception of β-blockers in those with one diagnosis. However, the disparity in AHM MPRs in donors compared with nondonors dissipated after increased numbers of clinical hypertension diagnoses, such among those with at least 3 hypertension diagnoses, use of all AHM classes except diuretics was similar among donors and non-donors.
Hypertension and related care are intermediate health outcomes warranting better understanding after live kidney donation. We examined a linkage of OPTN registry data for living kidney donors with pharmacy claims from a private health insurer to address knowledge gaps in requirements for pharmacologic hypertension treatments after donation. Based on this unique information source,
Antihypertensive Medication Use in Living Kidney Donors
Am J Nephrol 2014;40:174–183 DOI: 10.1159/000365157
Discussion
179
Downloaded by: Univ. of California Santa Barbara 198.143.33.33 - 8/29/2015 12:43:21 PM
0
Risk perspective
Metric
Examples
Addressed in current study
Descriptive
Frequency of events after donation
Frequency of AHM use after donation
X
Comparative, within-donor
Relative outcomes in donor subgroups
Differences in AHM use among African-American vs. White living donors
X
Comparative, donor vs. general non-donor
Relative outcomes in donors vs. general experience (often demographically matched, but not screened for baseline health status)
Differences in AHM use among living donors vs. general population non-donors
X
Attributable, donor vs. highly selected nondonor
Relative outcomes in donors vs. persons who would otherwise meet donor selection criteria (designed to simulate counterfactual experience of life without donation)
Differences in AHM use among living donors vs. non-donors selected for baseline good health
we observed several key findings: (1) Overall, 17.8% of the donor sample filled at least one AHM by 5 years postdonation. (2) Within the living donor sample, AfricanAmericans had 37% higher likelihood of post-donation AHM fills than White donors after adjustment for age and sex. Racial differences were especially strong for diuretics and vasodilators, but also present for ACEi/ARBs and calcium channel blockers. (3) The likelihood of postdonation AHM fills increased with baseline age at donation. (4) As an explicitly defined ‘general population’ comparison, AHM exposure (quantified by MPRs) was lower among donors than age- and sex-matched general non-donors in the same insurance plan. (5) Among donors with clinical hypertension diagnoses, AHM treatment exposure rose after multiple reported diagnoses, suggesting conversion from initial observation/lifestyle management to pharmaceutical treatment. (6) AHM exposure was also lower among donors than among nondonors with hypertension, but the disparity decreased and persisted only for diuretics in patients with at least 3 hypertension diagnoses. Racial variation in the frequency of hypertension after living donation has been recently recognized. While hypertension was identified in 25% of a cohort of 255 White donors from one center assessed at an average of 12 years after donation [22], by comparison, notably higher estimates of hypertension in 41% of African-American donors at earlier average assessment times of 7 years postdonation were found in two small cohorts [12, 23]. Based on linkage of OPTN registry data with private and public insurance medical claims, we found that, as compared with White donors, African-American donors had 40– 50% increased relative risks of post-donation hyperten180
Am J Nephrol 2014;40:174–183 DOI: 10.1159/000365157
sion diagnoses, and that racial differences were particularly strong for coded malignant hypertension [10, 11]. Our current study confirms these recent observations, here by using pharmacy claims for AHM (a measure of treated hypertension) as the outcome measure. While AHMs may be used for other conditions aside from hypertension (e.g. edema or cardiac conditions), our estimate of 37% higher relative likelihood of AHM use in African-American compared with White donors is very similar to the relative differences estimated using diagnostic claims for hypertension [10, 11]. Consistent with established associations of older age with hypertension risk in the general population as well as with hypertension diagnoses after kidney donation [10, 11], older age at donation was associated with increased likelihood of use of all AHM classes after donation in the current study. In most contemporary societies, blood pressure and requirements for pharmacologic treatment of hypertension rise with aging [24]. In addition to associations with baseline age, AHM use was higher among donors sampled later compared with earlier after donation in the current study, suggesting increased AHM requirements over time after donation. Determining how kidney donation modifies expected aging-related requirements for AHM warrants further study. Perspectives of risk among living donors include descriptions of event frequencies after donation, within-donor comparisons across donor subgroups (e.g. outcomes variation according to donor race), comparisons of outcomes among donors versus a general population, and when data for control screening are available, comparisons to non-donors selected for baseline good health (table 2). Recognition of the type of comparison is critical Lentine/Schnitzler/Garg/Xiao/Axelrod/ Tuttle-Newhall/Brennan/Segev
Downloaded by: Univ. of California Santa Barbara 198.143.33.33 - 8/29/2015 12:43:21 PM
Table 2. Perspectives of risk among living donors
overall, 18% of living donors in a recent US sample lacked insurance at donation, and insurance access varied demographically, such that 30–40% of young, African-American male donors were uninsured. Further, follow-up reporting deficiencies by centers are greater in these groups [6]. A commitment to follow-up and basic healthcare access is critical so that donors who have or develop conditions such as hypertension can be recognized and treated. Limitations of the current study include factors related to the sample and outcome measures. The outcomes measures were derived from insurance data, and uninsured living donors are not captured. Electronic pharmacy claims and fill records have been shown to be highly accurate records of physician prescribing [28–30], but claims data do not capture use of over-the-counter medications. Claims data are not medical records, and we lacked information on non-pharmacological treatments for hypertension such as lifestyle modifications. The MPR metric used in secondary analyses is an established measure of medication exposure that has been applied to combine AHM use from diverse classes into a single parameter, but does not include dose information given the diversity of agents aggregated within the metric [19–21]. Pre-donation insurance benefits were captured for only a minority of the donors (7.7%), and thus, information on pre-donation diagnoses was not adequate for inclusion. The OPTN began collecting information on pre-donation hypertension in June 2004: 12 of 399 donors from June 2004 through 2007 in our sample had OPTN reports of predonation hypertension, of whom 11 were White, 1 was Hispanic, and 0 were African-American – while this subgroup representing 3% of the full sample was too small for analysis, these patterns suggest that preexisting hypertension was not the cause of higher post-donation AHM requirements in African-American donors. Given the duration of captured insurance benefits, we did not have sufficient data to screen controls for baseline comorbidity, and thus the donor versus control comparisons must be interpreted with the limitations of other general comparisons (table 2) [10, 22, 25, 26]. Race information was not available for non-donors; however, as the representation of African-Americans as 13% of the study donor sample is similar to the proportion of AfricanAmericans in the general US population and we included matching of donors and controls by hypertension type, African-Americans representation is unlikely to have produced substantial overestimates of AHM requirements in the non-donors. The limited scope and 2-year duration of national registry follow-up is inadequate to characterize the long-
Antihypertensive Medication Use in Living Kidney Donors
Am J Nephrol 2014;40:174–183 DOI: 10.1159/000365157
181
Downloaded by: Univ. of California Santa Barbara 198.143.33.33 - 8/29/2015 12:43:21 PM
for interpretation. The data available in the current study were suitable for addressing the first 3 perspectives of risk, with a primary focus on within-donor comparisons of AMH fills according to donor race. Use of samples from the general population contextualizes donor outcomes in relation to average population health and such comparisons are commonly reported [10, 22, 25, 26]. In the current study, we included secondary analyses comparing AHM MPRs among donors versus age- and sex-matched general insurance beneficiaries. While by definition the study design does not address the direct impact of donation itself on requirements for AHM, these data frame AHM use among donors against use in an average population and demonstrate that AHM requirements after donation do not exceed general experience. Notably, two recent studies found increased likelihoods of hypertension diagnoses in donors compared with samples screened for baseline good health, including among a small sample of African-American donors [9, 12]. Efforts to assemble medication fill data among racially diverse samples screened for baseline health comparable to selected donors are warranted to define the direct impact of donation on AHM requirements. In additional secondary analyses, we examined postdonation AHM use in samples with diagnosed hypertension. AHM use increased after more frequently reported clinical hypertension diagnoses, suggesting conversion from initial observation to pharmacologic therapy. Further, while use of any AHM and most classes was lower in donors compared with non-donors matched for age at first hypertension diagnosis, sex, and coded hypertension type, the disparity in AHM MPRs in donors compared with non-donors dissipated after increased numbers of clinical hypertension diagnoses, such that treatment patterns in donors with ≥3 hypertension diagnoses resembled that of general non-donors. Classification of hypertension as benign or malignant in claims data is performed by the reporting provider and future study using claims integrated with records of measured blood pressure values is needed to define the correlation of medical coding with actual blood pressure levels. Nonetheless, our findings identify donors with multiple hypertension diagnoses as a group requiring substantial AHM treatment and deserving close monitoring for long-term sequelae of hypertension. Demonstration of pharmaceutical care needs of insured donors to manage important comorbidities, regardless of direct impact of donation itself, supports the need for long-term post-donation follow-up and access to healthcare for all donors. Gibney et al. [27] reported that,
term health of US living donors [4–6]. Innovative approaches to capturing outcomes among representative, diverse samples of living donors are needed. Supplementing the national registry with secondary data sources was endorsed by a consensus conference convened to evaluate ‘Living Kidney Donor Follow-Up: State-of-the-Art and Future Directions’ [31] and by individual experts in living donor care [32]. Methodologically, the current study demonstrates the value of pharmacy claims to describe an understudied dimension of health after living kidney donation. Integrated registry and pharmacy claims data can expand knowledge from other sources on treatment patterns after donation in a relatively efficient manner, without imposing time or participation burdens on donors and centers [33]. In conclusion, based on linkage of national transplant registry and pharmacy claims data, we found that while use of AHM after kidney donation is lower than among unscreened, age- and sex-matched general persons, there is racial variation in post-donation AHM use. This variation includes more common use of diuretics, ACEi/ARBs, calcium channel blockers, and vasodilators in African-American compared with White donors. Post-donation AHM use increases after more frequently reported clinical hypertension diagnoses, suggesting conversion from initial observation to phar-
macologic therapy, and treatment in donors with multiple hypertension diagnoses resembles that of general non-donors. Integrated registry and pharmacy records provide a novel tool for pharmacoepidemiologic investigations of delivered healthcare in living donors. Recognition of pharmaceutical care needs to manage important comorbidities among insured donors supports the need for long-term follow-up and healthcare access for all living donors.
Acknowledgements This work was supported by a grant from the National Institutes of Health (NIH)/National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) R01-DK096008. The data reported here have been supplied by UNOS as the contractor for the OPTN. The interpretation and reporting of these data are the responsibility of the authors and in no way should be seen as an official policy of or interpretation by UNOS, the SRTR, the US Government, or the National Institutes of Health. An abstract describing portions of this work was presentated at the World Transplant Congress, July 2014, San Francisco, Calif., USA.
Disclosure Statement The authors have no conflicts of interest to disclose.
1 OPTN – Organ Procurement and Transplantation Network/UNOS (United Network for Organ Sharing): National data reports, living donors, latest data. http://optn.transplant. hrsa.gov/latestData/rptData.asp (accessed September 12, 2012). 2 Lentine KL, Segev DL: Health outcomes among non-Caucasian living kidney donors: knowns and unknowns. Transpl Int 2013;26:853. 3 Taler SJ, Messersmith EE, Leichtman AB, Gillespie BW, Kew CE, Stegall MD, et al: Demographic, metabolic, and blood pressure characteristics of living kidney donors spanning five decades. Am J Transplant 2013; 13: 390– 398. 4 OPTN – Organ Procurement and Transplantation Network/UNOS (United Network for Organ Sharing): OPTN policies; policy 14: living donation. http://optn.transplant.hrsa. gov/ContentDocuments/OPTN_Policies.pdf (accessed February 7, 2014). 5 Wainright J: Living Donor Follow-Up Metrics, 2012. Richmond/VA, United Network for Organ Sharing, 2012. 6 Ommen ES, LaPointe Rudow D, Medapalli RK, Schroppel B, Murphy B: When good intentions are not enough: obtaining follow-up
182
7
8
9
10
11
data in living kidney donors. Am J Transplant 2011;11:2575–2581. United States Renal Data System. 2013 Annual Data Report. Chapter 1: Incidence, prevalence, patient characteristics, and treatment modalities. Figure 1.7: Incidence, counts, and adjusted rates of ESRD, by primary diagnosis. http://www.usrds.org/2013/pdf/v2_ch1_13. pdf (accessed April 16, 2014). Boudville N, Prasad GV, Knoll G, Muirhead N, Thiessen-Philbrook H, Yang RC, et al: Meta-analysis: risk for hypertension in living kidney donors. Ann Intern Med 2006; 145: 185–196. Garg AX, Prasad GV, Thiessen-Philbrook HR, Ping L, Melo M, Gibney EM, et al: Cardiovascular disease and hypertension risk in living kidney donors: an analysis of health administrative data in Ontario, Canada. Transplantation 2008;86:399–406. Lentine KL, Schnitzler MA, Xiao H, Saab G, Salvalaggio PR, Axelrod D, et al: Racial variation in medical outcomes among living kidney donors. N Engl J Med 2010;363:724– 732. Lentine KL, Schnitzler MA, Xiao H, Axelrod D, Garg AX, Tuttle-Newhall JE, et al: Consis-
Am J Nephrol 2014;40:174–183 DOI: 10.1159/000365157
12
13
14
15
tency of racial variation in medical outcomes among publicly and privately insured living kidney donors. Transplantation 2014;97:316– 324. Doshi MD, Goggins MO, Li L, Garg AX: Medical Outcomes in African American Live Kidney Donors: A Matched Cohort Study. Am J Transplant 2013;13:111–118. Muzaale AD, Massie AB, Wang MC, Montgomery RA, McBride MA, Wainright JL, et al: Risk of end-stage renal disease following live kidney donation. JAMA 2014; 311: 579– 586. Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JL Jr, et al: The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 Report. JAMA 2003; 289: 2560–2572. James PA, Oparil S, Carter BL, Cushman WC, Dennison-Himmelfarb C, Handler J, et al: 2014 evidence-based guideline for the management of high blood pressure in adults: report from the panel members appointed to the Eighth Joint National Committee (JNC 8). JAMA 2014;311:507–520.
Lentine/Schnitzler/Garg/Xiao/Axelrod/ Tuttle-Newhall/Brennan/Segev
Downloaded by: Univ. of California Santa Barbara 198.143.33.33 - 8/29/2015 12:43:21 PM
References
Antihypertensive Medication Use in Living Kidney Donors
21 Roy L, White-Guay B, Dorais M, Dragomir A, Lessard M, Perreault S: Adherence to antihypertensive agents improves risk reduction of end-stage renal disease. Kidney Int 2013; 84: 570–577. 22 Ibrahim HN, Foley R, Tan L, Rogers T, Bailey RF, Guo H, et al: Long-term consequences of kidney donation. N Engl J Med 2009; 360: 459–469. 23 Nogueira JM, Weir MR, Jacobs S, Haririan A, Breault D, Klassen D, et al: A study of renal outcomes in African-American living kidney donors. Transplantation 2009;88:1371–1376. 24 Vasan RS, Beiser A, Seshadri S, Larson MG, Kannel WB, D’Agostino RB, et al: Residual lifetime risk for developing hypertension in middle-aged women and men: The Framingham Heart Study. JAMA 2002;287:1003–1010. 25 Fehrman-Ekholm I, Elinder CG, Stenbeck M, Tyden G, Groth CG: Kidney donors live longer. Transplantation 1997;64:976–978. 26 Mjoen G, Reisaeter A, Hallan S, et al: Overall and cardiovascular mortality in Norwegian kidney donors compared to the background population. Nephrol Dial Transplant 2012; 27:443–447. 27 Gibney EM, Doshi MD, Hartmann EL, Parikh CR, Garg AX: Health insurance status of US living kidney donors. Clin J Am Soc Nephrol 2010;5:912–916.
28 Levy AR, O’Brien BJ, Sellors C, Grootendorst P, Willison D: Coding accuracy of administrative drug claims in the Ontario Drug Benefit database. Can J Clin Pharmacol 2003;10:67–71. 29 Lau HS, de Boer A, Beuning KS, Porsius A: Validation of pharmacy records in drug exposure assessment. J Clin Epidemiol 1997; 50: 619–625. 30 Boethius G: Recording of drug prescriptions in the county of Jamtland, Sweden. II. Drug exposure of pregnant women in relation to course and outcome of pregnancy. Eur J Clin Pharmacol 1977;12:37–43. 31 Leichtman A, Abecassis M, Barr M, Charlton M, Cohen D, Confer D, et al: Living kidney donor follow-up: state-of-the-art and future directions, conference summary and recommendations. Am J Transplant 2011;11:2561–2568. 32 Gaston RS, Young CJ: Living donor nephrectomy: understanding long-term risk in minority populations. Am J Transplant 2010;10: 2574–2576. 33 Lentine KL, Segev DL: Better understanding live donor risk through big data. Clin J Am Soc Nephrol 2013;8:1645.
Am J Nephrol 2014;40:174–183 DOI: 10.1159/000365157
183
Downloaded by: Univ. of California Santa Barbara 198.143.33.33 - 8/29/2015 12:43:21 PM
16 Lentine KL, Anyaegbu E, Gleisner A, Schnitzler MA, Axelrod D, Brennan DC, et al: Understanding medical care of transplant recipients through integrated registry and pharmacy claims data. Am J Nephrol 2013; 38: 420–429. 17 Buchanan PM, Schnitzler MA, Brennan DC, Dzebisashvili N, Willoughby LM, Axelrod D, et al: Novel methods for tracking longterm maintenance immunosuppression regimens. Clin J Am Soc Nephrol 2008; 3: 117– 124. 18 Lentine KL, Villines TC, Xiao H, Schnitzler MA, Brennan DC, Abbott KC, et al: Cardioprotective medication use after acute myocardial infarction in kidney transplant recipients. Transplantation 2011;91:1120–1126. 19 Pinsky BW, Takemoto SK, Lentine KL, Burroughs TE, Schnitzler MA, Salvalaggio PR: Transplant outcomes and economic costs associated with patient noncompliance to immunosuppression. Am J Transplant 2009; 9: 2597–2606. 20 Osterberg L, Blaschke T. Adherence to medication. N Engl J Med 2005;353:487–497.
Journal of
Nephrology
Received: May 17, 2014 Accepted: June 7, 2014 Published online: September 2, 2014
Am J Nephrol 2014;40:174–183 DOI: 10.1159/000365157
Understanding Antihypertensive Medication Use after Living Kidney Donation through Linked National Registry and Pharmacy Claims Data Krista L. Lentine a, b Mark A. Schnitzler a, b Amit X. Garg f Huiling Xiao a David Axelrod d Janet E. Tuttle-Newhall b Daniel C. Brennan c Dorry L. Segev e
a
Center for Outcomes Research and b Division of Abdominal Transplantation, Department of Surgery, Saint Louis University School of Medicine, c Division of Nephrology, Department of Medicine, Washington University School of Medicine, St. Louis, Mo., d Division of Abdominal Transplantation, Department of Surgery, Dartmouth Hitchcock Medical Center, Hanover, N.H., and e Division of Abdominal Transplantation, Department of Surgery, Johns Hopkins University, Baltimore, Md., USA; f Division of Nephrology, Western University, London, Ont., Canada
Abstract Background: Use of antihypertensive medications (AHM) after living kidney donation is not well described. Methods: We examined a database wherein national transplant registry data for 4,650 living kidney donors in 1987–2007 were linked to pharmacy claims from a US private health insurer (2000–2007 claims) to identify post-donation AHM fills. Cox regression with left- and right-censoring was used to estimate the frequencies and relative likelihood (adjusted hazards ratios, aHR) of post-donation AHM fills according to donor demographic traits. Medication possession ratio (MPRs), defined as (days of AHM dispensed)/(days observed), were also compared among donors and non-donor general beneficiaries. Results: Overall, 17.8% of the sample filled at least one AHM by 5 years post-donation. As compared with White living donors, African-Americans had 37% higher relative likelihood of any AHM use after donation (aHR 1.37, p < 0.0007), including significantly higher likelihoods of filling
© 2014 S. Karger AG, Basel 0250–8095/14/0402–0174$39.50/0 E-Mail [email protected] www.karger.com/ajn
diuretics (aHR 2.25, p < 0.0001), ACEi/ARBs (aHR 1.46, p < 0.01), calcium channel blockers (aHR 1.56, p = 0.03), and vasodilators/other agents (aHR 2.17, p = 0.03). MPRs for any AHM and subcategories were lower among donors compared with age- and sex-matched non-donors. However, AHM MPRs rose in donors with multiple hypertension diagnoses, and prescription fill exposure for all AHM classes except diuretics was similar among donors and general nondonors with ≥3 hypertension diagnoses. Conclusions: While AHM requirements are lower after kidney donation than among unscreened general persons, racial variation in AHM use occurs in privately insured donors. Demonstration of pharmaceutical care needs of insured donors supports the need for long-term follow-up and healthcare access for all donors. © 2014 S. Karger AG, Basel
Introduction
In the context of the organ shortage, kidney transplantation from living donors has increased markedly over the last several decades [1] and has evolved to more commonly include donors with baseline medical complexity Krista L. Lentine, MD, PhD Saint Louis University, Salus Center 4th Floor 3545 Lafayette Avenue, St. Louis, MO 63104 (USA) E-Mail lentinek @ slu.edu
Downloaded by: Univ. of California Santa Barbara 198.143.33.33 - 8/29/2015 12:43:21 PM
Key Words Antihypertensive agents · Hypertension · Insurance · Kidney transplantation · Living donors · Pharmacy claims · Registries
and greater racial and ethnic diversity [2, 3]. Importantly, despite the growth and expansion of living donation, mandated follow-up of living donors in the USA has been limited in scope, duration and completeness. Currently, the Organ Procurement and Transplantation Network (OPTN) collects data on living donors for only 2 years of follow-up [4], and incomplete reporting and donor loss to follow-up have been common [5, 6]. While OPTN/ United Network for Organ Sharing (UNOS) policies were modified in 2013 to include mandated thresholds for collection of follow-up information by centers, compliance with these thresholds could still leave missing data (up to 20% missing for clinical and 30% for laboratory results) [4], and the predictive value of information at 2 years for important dimensions of long-term donor health is uncertain. Hypertension is an intermediate health outcome warranting better understanding after live kidney donation. Given established associations of hypertension as a leading cause of chronic kidney disease and end-stage renal disease (ESRD) in the general population [7], there is theoretical concern that post-donation hypertension may accelerate glomerular filtration rate decline in donors. Hypertension and renal function are reciprocally related, and data from predominantly White cohorts suggest increased hypertension risk in prior donors compared to the general population, possibly due to physiological alterations (hyperfiltration in the remaining kidney, changes in vascular tone and renin-angiotensin-aldosterone regulation) and/or heightened follow-up [8, 9]. Based on integration of the OPTN registry and administrative billing claims from public and private insurers, we recently found racial variation in post-donation hypertension diagnoses, with 40–50% higher relative risks of hypertension diagnoses in African-American versus White donors, including two to three times the relative risks of malignant hypertension [10, 11]. A recent report of 103 African-American donors at two centers suggested that the frequency of post-donation hypertension may exceed that of matched controls, noting a high proportion of previously undiagnosed hypertension identified through study encounters [12]. Further, a new study of data for 96,217 US living donors identified a small but significant increase in ESRD risk attributable to donation that was highest in African-Americans [13], illustrating the need to understand mediators of disparities in post-donation ESRD. Management of hypertension may include lifestyle modification and close observation, followed by pharmacologic therapy with antihypertensive medications (AHM) in patients who do not achieve treatment
goals [14, 15]. To date, limited information has been reported on AHM use in demographically diverse donor samples. Pharmacy claims offer a non-obtrusive measure of prescribed healthcare that do not rely on patient self-report and are increasingly used in observational investigations of large populations including transplant-related epidemiologic studies [16–21]. To advance understanding of medically treated hypertension after live kidney donation, we examined a linkage of OPTN registry data for living donors with administrative records from a private health insurer that includes pharmacy claims for prescription medications. Our primary aim was to identify post-donation AHM fills among a large sample of US living donors, overall and by agent class, and to compare post-donation AHM fills within donors according to race. In secondary analyses, we also compared AHM exposure among donors to that in age- and sex-matched general non-donors overall, as well as fill patterns among donors and non-donors with clinical hypertension diagnoses.
Antihypertensive Medication Use in Living Kidney Donors
Am J Nephrol 2014;40:174–183 DOI: 10.1159/000365157
Data Source and Sample This study used data from the OPTN. The OPTN data system includes information on all donors, waitlisted candidates, and transplant recipients in the USA, submitted by the members of the OPTN. The Health Resources and Services Administration (HRSA), US Department of Health and Human Services provides oversight to the activities of the OPTN contractor. Study data were assembled by linking OPTN records for living kidney donors with administrative data from a national private health insurer. After approval by HRSA and the Saint Louis University Institutional Review Board, beneficiary identifier numbers from the insurer’s electronic databases were linked using names and birthdates to unique OPTN identifiers for living kidney donors. Analyses were performed using Health Information Portability and Accountability Act-compliant, limited datasets with all direct identifiers removed. We included living kidney donors who had records of donating between October 1987 and July 2007 and benefits under the participating insurer after donor nephrectomy at some point in May 2000 to December 2007 (the period of available claims data). All study participants were simultaneously enrolled in medical and pharmacy benefits with this insurer exclusively during the study period. Because of the large sample size, the anonymity of the patients studied, and the non-intrusive nature of the research, a waiver of informed consent was granted per the Department of Health and Human Services Code of Federal Regulations (Title 45, Part 46, Paragraph 46.116). Non-donors were sampled from records of general insurance beneficiaries enrolled in the same plan at some point during May 2000 to December 2007.
175
Downloaded by: Univ. of California Santa Barbara 198.143.33.33 - 8/29/2015 12:43:21 PM
Methods
Statistical Analyses Correlates of AHM Use after Living Kidney Donation Datasets were merged and analyzed with SAS for Windows software, version 9.3 (SAS Institute, Inc., Cary, N.C., USA). As windows of captured insurance benefits varied across the sample, Cox regression with left- and right-censoring was used to estimate the frequency of patients with AMH fills over time after donation, and associations (adjusted hazards ratios, aHR) between donor traits (particularly race) and AHM fills. Censoring was applied from donation to insurance enrollment and after the end of an individual’s captured insurance benefits (or end of study, December 2007). Comparison of AHM MPRs in All Donors and General Non-Donors To compare AHM exposure after donation to use among a non-donor sample as a secondary analysis, living donors were matched one-to-one with general insurance beneficiaries by sex and age when benefits began. Maximum observation time, defined by benefits duration in each matched pair, was limited to the shortest available in the pair. We compared MPRs for any AHM and AHM classes in donors and matched general non-donors using the paired Wilcoxon signed-rank test. A schematic of the design for subject matching and AHM MPR computation among the full donor cohort and age- and sex-matched non-donors is shown in figure 1a. Comparison of AHM MPRs in Donors and General Non-Donors with Hypertension We also sought to describe AHM exposure among living donors and non-donors with clinical hypertension diagnoses. Hypertension diagnoses were ascertained from medical billing claims with corresponding International Classification of Disease, Ninth
176
Am J Nephrol 2014;40:174–183 DOI: 10.1159/000365157
Revision, Clinical Modification (ICD-9-CM) diagnosis codes and subclassified as benign (401.1, 402.10, 402.11, 403.10, 403.11, 404.10–404.13, 405.11, 405.19), malignant (401.0, 402.00, 402.01, 403.00, 403.01, 404.00–404.03, 405.01, 405.09) or unspecified (401.9, 402.90, 402.91, 403.90, 403.91, 404.90–404.93, 405.91, 405.99), as previously described [11]. Living donors with hypertension were matched to non-donors with hypertension by age at hypertension diagnosis, sex, and hypertension type (fig. 1b). Here observation time for MPR computation began at the first captured hypertension diagnosis. Finally, we examined AHM exposure patterns among living donors with multiple post-donation hypertension diagnoses as an index of severity by repeating matching procedures and AHM MPR computations among donors and nondonors with ≥2 and ≥3 captured hypertension diagnoses. Observation time for AHM MPR computation in these analyses began at the date of the second and third captured hypertension diagnosis, respectively.
Results
Demographic Correlates of AHM Use after Living Kidney Donation Characteristics of the 4,650 live kidney donors in the study cohort have been previously reported [10]. Among the sample, 13.1% were African-American, 76.3% White, 8.2% Hispanic, and 2.4% other races (table 1). Mean age at donation was 37.2 ± 10 years. The median times from donation to the start and end of observed insurance eligibility were 4.9 and 7.7 years, respectively. Distributions of race and sex in the linked donor sample were similar to that of all living kidney donors in the OPTN in the period [10]. Overall, at 5 years post-donation, 17.8% (95% CI 15.1– 20.0) of the sample filled prescriptions for at least one AHM, including fills for diuretics in 7.2% (95% CI 5.5– 8.9) and ACEi/ARBs in 5.9% (95% CI 4.4–7.4). By 5 years, 6.6% (95% CI 5.0–8.3) of the donor sample had received β-blockers, 3.4% (95% CI 2.2–4.6) had filled calcium channel blocker prescriptions, and 1.2% (95% CI 0.05– 1.9) had received vasodilators/other agents. In multivariate regression including adjustment for donor age and sex, African-American race was associated with 37% higher likelihood of fills for any AHM after donation (aHR 1.37, 95% CI 1.09–1.72) (table 1). Compared with White donors, African-American donors also had more than twice the likelihoods of filling diuretics (aHR 2.25, 95% CI 1.71–2.95) and vasodilators/other agents (aHR 2.17, 95% CI 1.07–4.42), as well as significantly higher likelihood of filling ACEi/ARBs (aHR 1.46, 95% CI 1.08–1.99) and calcium channel blockers (aHR 1.58, 95% CI 1.03–2.42) after donation. Lentine/Schnitzler/Garg/Xiao/Axelrod/ Tuttle-Newhall/Brennan/Segev
Downloaded by: Univ. of California Santa Barbara 198.143.33.33 - 8/29/2015 12:43:21 PM
Definitions of Outcomes and Covariates The primary outcome of interest was pharmacy claims for AHM fills after donation. AHM were also subcategorized as ACE inhibitors/angiotensin II receptor blockers (ACEi/ARBs), βblockers, calcium channel blockers, diuretics or vasodilators/other agents. Combination pills (e.g. diuretics with other AHM agents) were considered to contribute mediation to each of the component classes. Primary analyses considered time to first fill of any AHM or AHM class, respectively, among the donor sample. In secondary analyses, we computed medication possession ratios (MPRs), a metric quantifying the fraction of days of captured insurance enrollment for which AHM were prescribed [19– 21]. MPRs were defined as: [days of medication supplied over an observation window]/[days of observation], where observation windows were defined as the period of captured insurance benefits for an individual. To account for concomitant use of multiple agents, the MPR metric for any AHM exposure aggregated fill days from different classes even if prescription dates were overlapping, such that the maximum possible MPR for any AHM may exceed 1.0. Demographic data from the OPTN at donor nephrectomy included age, gender, and race as reported by the transplant center. The insurance records include information on age and sex, but not race; thus race information was not available for non-donors. Body mass index was reported to the OPTN for only 5.3% of the linked donors and so was inadequate for analysis.
Color version available online
Match by: Age at INS enrollment, sex Non-donor INS benefits AHM MPR
Days fill
Days fill
Days fill
Days fill
Days fill
Days of AHM Observation days
Days of AHM Observation days
LKD INS benefits
a
Start observation at INS enrollment (at or after donation)
Limit observation to minimum in matched pair
Match by: Age at HTN Dx, sex, HTN type Non-donor INS benefits AHM MPR
Days fill
Days fill
HTN Dx in medical claims Days fill
Fig. 1. Schematic of design for AHM medi-
Days fill
Days of AHM Observation days
LKD INS benefits
b
Start observation at HTN Dx
Limit observation to minimum time after HTN Dx in matched pair
AHM MPRs in All Donors and General Non-Donors AHM exposure in donors and age- and sex-matched general non-donors was compared using MPRs, the fraction of days of captured insurance enrollment for which AHM were prescribed. Overall, AHM MPRs were lower among the full donor sample than among age- and sexmatched non-donors: any AHM, 11 vs. 26%, p < 0.00001; diuretics, 3 vs. 8%, p < 0.0001; ACEi/ARBs, 4 vs. 11%, p < 0.0001; β-blockers, 3 vs. 6%; calcium channel blockers, 2 vs. 4%, and vasodilators/other agents, 1 vs. 0%. Stratify-
ing the full donor cohort by time from nephrectomy to insurance enrollment demonstrated higher MPRs among live donors enrolled after, compared with before, the median time to enrollment: any AHM, 18 vs. 5%; diuretics, 5 vs. 1%; ACEi/ARBs, 6 vs. 2%; β-blockers, 5 vs. 1%, calcium channel blockers, 2 vs. 1%, and vasodilators/other agents, 1 vs. 0% (fig. 2). However, MPRs remained lower in donors compared with their age- and sex-matched non-donor controls after stratification by enrollment timing.
Antihypertensive Medication Use in Living Kidney Donors
Am J Nephrol 2014;40:174–183 DOI: 10.1159/000365157
177
Downloaded by: Univ. of California Santa Barbara 198.143.33.33 - 8/29/2015 12:43:21 PM
cation possession ratio computation among (a) the full living donor cohort and ageand sex-matched general non-donors and (b) donors and non-donors with hypertension diagnoses, matched by age at diagnosis, sex and hypertension type. AHM = Antihypertensive medication; Dx = diagnosis; HTN = hypertension; INS = insurance plan; LKD = living kidney donor; MPR = medication possession ratio.
Days fill
Days of AHM Observation days
LKDs General controls
MPR (%)
40 30 20
*
10 0
*
* Any
Diuretics
ACEi/ARBs
*
*
BBs
CCBs
†† Others
LKDs enrolled after 4.9 years post-donation
50 40 MPR (%)
Fig. 2. AHM medication possession ratios
among living donors compared with ageand sex-matched general non-donor controls, stratified by time from donation to study enrollment as below or above the sample median. ACEi/ARBs = ACE inhibitors/angiotensin II receptor blockers; BBs = β-blockers; CCBs = calcium channel blockers; LKD = living kidney donor; MPR = medication possession ratio. † p = 0.05; ‡ p < 0.05–0.0001; * p < 0.0001.
Color version available online
LKDs enrolled within 4.9 years post-donation
50
30
*
20 10 0
*
* Any
Diuretics
* ACEi/ARBs
BBs
CCBs
† Others
Table 1. Adjusted associations of baseline demographic factors with likelihoods of post-donation pharmacy fills for any AMH and AHM
subclasses in privately insured living kidney donors Trait distributions Age at donation (per year) Male gender Race White, Non-Hispanic African-American Hispanic Other
any AHM
diuretics
ACEi/ARBs
CCBs
BBs
vasodilators/ others
37.2±10.0# 1.05 (1.04–1.06)§ 1.04 (1.03–1.05)§ 1.06 (1.05–1.07)§ 1.05 (1.04–1.07)§ 1.06 (1.04–1.07)§ 1.06 (1.03–1.09)§ 45.4% 0.95 (0.82–1.11) 0.63 (0.50–0.78)§ 1.30 (1.05–1.61)* 0.83 (0.61–1.13) 1.13 (0.87–1.46) 1.07 (0.61–1.87) 76.3% 13.1% 8.2% 81.2%
reference 1.37 (1.09–1.72)* 1.02 (0.74–1.40) 0.47 (0.21–1.05)
reference 2.25 (1.71–2.95)§ 1.07 (0.68–1.70) 0.37 (0.09–1.51)
reference 1.46 (1.08–1.99)* 1.18 (0.77–1.81) 0.52 (0.17–1.64)
reference 1.58 (1.03–2.42)* 0.82 (0.40–1.68) 0.72 (0.18–2.92)
reference 1.07 (0.71–1.61) 0.74 (0.40–1.37) 0.22 (0.03–1.57)
reference 2.17 (1.07–4.42)* 0.36 (0.052.66) 1.16 (0.16–8.49)
ACEi/ARBs = ACE inhibitors/angiotensin II receptor blockers; BBs = β-Blockers; CCBs = calcium channel blockers. # Mean ± SD; * p < 0.05–0.0001; ≤ 0.0001.
AHM MPRs in Donors and General Non-Donors with Hypertension MPRs computed among subgroups identified by the presence of clinical hypertension diagnoses were, as expected, substantially higher than among the full cohort overall. The MPR for any AHM exposure in donors with 178
Am J Nephrol 2014;40:174–183 DOI: 10.1159/000365157
at least one captured hypertension diagnosis was 61% (fig. 3). Further, there were graded increases in AHM MPRs among donors after increasing numbers of hypertension diagnoses, rising to 72% days exposed to any AHM in donors with at least two diagnoses and 90% in those with at least 3 diagnoses. MPRs for any AHM and Lentine/Schnitzler/Garg/Xiao/Axelrod/ Tuttle-Newhall/Brennan/Segev
Downloaded by: Univ. of California Santa Barbara 198.143.33.33 - 8/29/2015 12:43:21 PM
§ p
AHM pharmacy fill outcomes, aHR (95% CI)
0.9
Color version available online
Care after at least 1 HTN diagnosis
1.0
*
0.8
MPR
0.7 0.6 0.5 0.4
††
0.3
*
††
0.2 0.1 0
1.0 0.9
Any
Diuretics
ACEi/ARBs
BBs
CCBs
Others
Care after at least 2 HTN diagnoses
*
0.8 0.7 MPR
0.6 0.5
*
0.4
††
0.3 0.2
† †
††
BBs
CCBs
0.1
among living donors and general non-donors with clinical hypertension diagnoses, matched by age at hypertension diagnosis, sex and hypertension type (number per group by diagnosis frequency: ≥1, 894; ≥2, 672; ≥3, 500). ACEi/ARBs = ACE inhibitors/angiotensin II receptor blockers; BBs = β-blockers; CCBs = calcium channel blockers; HTN = hypertension; LKD = living kidney donor; MPR = medication possession ratio. ‡ p < 0.05–0.0001; * p < 0.0001.
MPR
Fig. 3. AHM medication possession ratios
1.1 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0
Any
Diuretics
ACEi/ARBs
Others
Care after at least 3 HTN diagnoses
*
LKDs General controls
††
Any
Diuretics
ACEi/ARBs
BBs
CCBs
Others
most classes remained lower in donors compared with non-donors matched for age at first hypertension diagnosis, sex and coded hypertension type, with the exception of β-blockers in those with one diagnosis. However, the disparity in AHM MPRs in donors compared with nondonors dissipated after increased numbers of clinical hypertension diagnoses, such among those with at least 3 hypertension diagnoses, use of all AHM classes except diuretics was similar among donors and non-donors.
Hypertension and related care are intermediate health outcomes warranting better understanding after live kidney donation. We examined a linkage of OPTN registry data for living kidney donors with pharmacy claims from a private health insurer to address knowledge gaps in requirements for pharmacologic hypertension treatments after donation. Based on this unique information source,
Antihypertensive Medication Use in Living Kidney Donors
Am J Nephrol 2014;40:174–183 DOI: 10.1159/000365157
Discussion
179
Downloaded by: Univ. of California Santa Barbara 198.143.33.33 - 8/29/2015 12:43:21 PM
0
Risk perspective
Metric
Examples
Addressed in current study
Descriptive
Frequency of events after donation
Frequency of AHM use after donation
X
Comparative, within-donor
Relative outcomes in donor subgroups
Differences in AHM use among African-American vs. White living donors
X
Comparative, donor vs. general non-donor
Relative outcomes in donors vs. general experience (often demographically matched, but not screened for baseline health status)
Differences in AHM use among living donors vs. general population non-donors
X
Attributable, donor vs. highly selected nondonor
Relative outcomes in donors vs. persons who would otherwise meet donor selection criteria (designed to simulate counterfactual experience of life without donation)
Differences in AHM use among living donors vs. non-donors selected for baseline good health
we observed several key findings: (1) Overall, 17.8% of the donor sample filled at least one AHM by 5 years postdonation. (2) Within the living donor sample, AfricanAmericans had 37% higher likelihood of post-donation AHM fills than White donors after adjustment for age and sex. Racial differences were especially strong for diuretics and vasodilators, but also present for ACEi/ARBs and calcium channel blockers. (3) The likelihood of postdonation AHM fills increased with baseline age at donation. (4) As an explicitly defined ‘general population’ comparison, AHM exposure (quantified by MPRs) was lower among donors than age- and sex-matched general non-donors in the same insurance plan. (5) Among donors with clinical hypertension diagnoses, AHM treatment exposure rose after multiple reported diagnoses, suggesting conversion from initial observation/lifestyle management to pharmaceutical treatment. (6) AHM exposure was also lower among donors than among nondonors with hypertension, but the disparity decreased and persisted only for diuretics in patients with at least 3 hypertension diagnoses. Racial variation in the frequency of hypertension after living donation has been recently recognized. While hypertension was identified in 25% of a cohort of 255 White donors from one center assessed at an average of 12 years after donation [22], by comparison, notably higher estimates of hypertension in 41% of African-American donors at earlier average assessment times of 7 years postdonation were found in two small cohorts [12, 23]. Based on linkage of OPTN registry data with private and public insurance medical claims, we found that, as compared with White donors, African-American donors had 40– 50% increased relative risks of post-donation hyperten180
Am J Nephrol 2014;40:174–183 DOI: 10.1159/000365157
sion diagnoses, and that racial differences were particularly strong for coded malignant hypertension [10, 11]. Our current study confirms these recent observations, here by using pharmacy claims for AHM (a measure of treated hypertension) as the outcome measure. While AHMs may be used for other conditions aside from hypertension (e.g. edema or cardiac conditions), our estimate of 37% higher relative likelihood of AHM use in African-American compared with White donors is very similar to the relative differences estimated using diagnostic claims for hypertension [10, 11]. Consistent with established associations of older age with hypertension risk in the general population as well as with hypertension diagnoses after kidney donation [10, 11], older age at donation was associated with increased likelihood of use of all AHM classes after donation in the current study. In most contemporary societies, blood pressure and requirements for pharmacologic treatment of hypertension rise with aging [24]. In addition to associations with baseline age, AHM use was higher among donors sampled later compared with earlier after donation in the current study, suggesting increased AHM requirements over time after donation. Determining how kidney donation modifies expected aging-related requirements for AHM warrants further study. Perspectives of risk among living donors include descriptions of event frequencies after donation, within-donor comparisons across donor subgroups (e.g. outcomes variation according to donor race), comparisons of outcomes among donors versus a general population, and when data for control screening are available, comparisons to non-donors selected for baseline good health (table 2). Recognition of the type of comparison is critical Lentine/Schnitzler/Garg/Xiao/Axelrod/ Tuttle-Newhall/Brennan/Segev
Downloaded by: Univ. of California Santa Barbara 198.143.33.33 - 8/29/2015 12:43:21 PM
Table 2. Perspectives of risk among living donors
overall, 18% of living donors in a recent US sample lacked insurance at donation, and insurance access varied demographically, such that 30–40% of young, African-American male donors were uninsured. Further, follow-up reporting deficiencies by centers are greater in these groups [6]. A commitment to follow-up and basic healthcare access is critical so that donors who have or develop conditions such as hypertension can be recognized and treated. Limitations of the current study include factors related to the sample and outcome measures. The outcomes measures were derived from insurance data, and uninsured living donors are not captured. Electronic pharmacy claims and fill records have been shown to be highly accurate records of physician prescribing [28–30], but claims data do not capture use of over-the-counter medications. Claims data are not medical records, and we lacked information on non-pharmacological treatments for hypertension such as lifestyle modifications. The MPR metric used in secondary analyses is an established measure of medication exposure that has been applied to combine AHM use from diverse classes into a single parameter, but does not include dose information given the diversity of agents aggregated within the metric [19–21]. Pre-donation insurance benefits were captured for only a minority of the donors (7.7%), and thus, information on pre-donation diagnoses was not adequate for inclusion. The OPTN began collecting information on pre-donation hypertension in June 2004: 12 of 399 donors from June 2004 through 2007 in our sample had OPTN reports of predonation hypertension, of whom 11 were White, 1 was Hispanic, and 0 were African-American – while this subgroup representing 3% of the full sample was too small for analysis, these patterns suggest that preexisting hypertension was not the cause of higher post-donation AHM requirements in African-American donors. Given the duration of captured insurance benefits, we did not have sufficient data to screen controls for baseline comorbidity, and thus the donor versus control comparisons must be interpreted with the limitations of other general comparisons (table 2) [10, 22, 25, 26]. Race information was not available for non-donors; however, as the representation of African-Americans as 13% of the study donor sample is similar to the proportion of AfricanAmericans in the general US population and we included matching of donors and controls by hypertension type, African-Americans representation is unlikely to have produced substantial overestimates of AHM requirements in the non-donors. The limited scope and 2-year duration of national registry follow-up is inadequate to characterize the long-
Antihypertensive Medication Use in Living Kidney Donors
Am J Nephrol 2014;40:174–183 DOI: 10.1159/000365157
181
Downloaded by: Univ. of California Santa Barbara 198.143.33.33 - 8/29/2015 12:43:21 PM
for interpretation. The data available in the current study were suitable for addressing the first 3 perspectives of risk, with a primary focus on within-donor comparisons of AMH fills according to donor race. Use of samples from the general population contextualizes donor outcomes in relation to average population health and such comparisons are commonly reported [10, 22, 25, 26]. In the current study, we included secondary analyses comparing AHM MPRs among donors versus age- and sex-matched general insurance beneficiaries. While by definition the study design does not address the direct impact of donation itself on requirements for AHM, these data frame AHM use among donors against use in an average population and demonstrate that AHM requirements after donation do not exceed general experience. Notably, two recent studies found increased likelihoods of hypertension diagnoses in donors compared with samples screened for baseline good health, including among a small sample of African-American donors [9, 12]. Efforts to assemble medication fill data among racially diverse samples screened for baseline health comparable to selected donors are warranted to define the direct impact of donation on AHM requirements. In additional secondary analyses, we examined postdonation AHM use in samples with diagnosed hypertension. AHM use increased after more frequently reported clinical hypertension diagnoses, suggesting conversion from initial observation to pharmacologic therapy. Further, while use of any AHM and most classes was lower in donors compared with non-donors matched for age at first hypertension diagnosis, sex, and coded hypertension type, the disparity in AHM MPRs in donors compared with non-donors dissipated after increased numbers of clinical hypertension diagnoses, such that treatment patterns in donors with ≥3 hypertension diagnoses resembled that of general non-donors. Classification of hypertension as benign or malignant in claims data is performed by the reporting provider and future study using claims integrated with records of measured blood pressure values is needed to define the correlation of medical coding with actual blood pressure levels. Nonetheless, our findings identify donors with multiple hypertension diagnoses as a group requiring substantial AHM treatment and deserving close monitoring for long-term sequelae of hypertension. Demonstration of pharmaceutical care needs of insured donors to manage important comorbidities, regardless of direct impact of donation itself, supports the need for long-term post-donation follow-up and access to healthcare for all donors. Gibney et al. [27] reported that,
term health of US living donors [4–6]. Innovative approaches to capturing outcomes among representative, diverse samples of living donors are needed. Supplementing the national registry with secondary data sources was endorsed by a consensus conference convened to evaluate ‘Living Kidney Donor Follow-Up: State-of-the-Art and Future Directions’ [31] and by individual experts in living donor care [32]. Methodologically, the current study demonstrates the value of pharmacy claims to describe an understudied dimension of health after living kidney donation. Integrated registry and pharmacy claims data can expand knowledge from other sources on treatment patterns after donation in a relatively efficient manner, without imposing time or participation burdens on donors and centers [33]. In conclusion, based on linkage of national transplant registry and pharmacy claims data, we found that while use of AHM after kidney donation is lower than among unscreened, age- and sex-matched general persons, there is racial variation in post-donation AHM use. This variation includes more common use of diuretics, ACEi/ARBs, calcium channel blockers, and vasodilators in African-American compared with White donors. Post-donation AHM use increases after more frequently reported clinical hypertension diagnoses, suggesting conversion from initial observation to phar-
macologic therapy, and treatment in donors with multiple hypertension diagnoses resembles that of general non-donors. Integrated registry and pharmacy records provide a novel tool for pharmacoepidemiologic investigations of delivered healthcare in living donors. Recognition of pharmaceutical care needs to manage important comorbidities among insured donors supports the need for long-term follow-up and healthcare access for all living donors.
Acknowledgements This work was supported by a grant from the National Institutes of Health (NIH)/National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) R01-DK096008. The data reported here have been supplied by UNOS as the contractor for the OPTN. The interpretation and reporting of these data are the responsibility of the authors and in no way should be seen as an official policy of or interpretation by UNOS, the SRTR, the US Government, or the National Institutes of Health. An abstract describing portions of this work was presentated at the World Transplant Congress, July 2014, San Francisco, Calif., USA.
Disclosure Statement The authors have no conflicts of interest to disclose.
1 OPTN – Organ Procurement and Transplantation Network/UNOS (United Network for Organ Sharing): National data reports, living donors, latest data. http://optn.transplant. hrsa.gov/latestData/rptData.asp (accessed September 12, 2012). 2 Lentine KL, Segev DL: Health outcomes among non-Caucasian living kidney donors: knowns and unknowns. Transpl Int 2013;26:853. 3 Taler SJ, Messersmith EE, Leichtman AB, Gillespie BW, Kew CE, Stegall MD, et al: Demographic, metabolic, and blood pressure characteristics of living kidney donors spanning five decades. Am J Transplant 2013; 13: 390– 398. 4 OPTN – Organ Procurement and Transplantation Network/UNOS (United Network for Organ Sharing): OPTN policies; policy 14: living donation. http://optn.transplant.hrsa. gov/ContentDocuments/OPTN_Policies.pdf (accessed February 7, 2014). 5 Wainright J: Living Donor Follow-Up Metrics, 2012. Richmond/VA, United Network for Organ Sharing, 2012. 6 Ommen ES, LaPointe Rudow D, Medapalli RK, Schroppel B, Murphy B: When good intentions are not enough: obtaining follow-up
182
7
8
9
10
11
data in living kidney donors. Am J Transplant 2011;11:2575–2581. United States Renal Data System. 2013 Annual Data Report. Chapter 1: Incidence, prevalence, patient characteristics, and treatment modalities. Figure 1.7: Incidence, counts, and adjusted rates of ESRD, by primary diagnosis. http://www.usrds.org/2013/pdf/v2_ch1_13. pdf (accessed April 16, 2014). Boudville N, Prasad GV, Knoll G, Muirhead N, Thiessen-Philbrook H, Yang RC, et al: Meta-analysis: risk for hypertension in living kidney donors. Ann Intern Med 2006; 145: 185–196. Garg AX, Prasad GV, Thiessen-Philbrook HR, Ping L, Melo M, Gibney EM, et al: Cardiovascular disease and hypertension risk in living kidney donors: an analysis of health administrative data in Ontario, Canada. Transplantation 2008;86:399–406. Lentine KL, Schnitzler MA, Xiao H, Saab G, Salvalaggio PR, Axelrod D, et al: Racial variation in medical outcomes among living kidney donors. N Engl J Med 2010;363:724– 732. Lentine KL, Schnitzler MA, Xiao H, Axelrod D, Garg AX, Tuttle-Newhall JE, et al: Consis-
Am J Nephrol 2014;40:174–183 DOI: 10.1159/000365157
12
13
14
15
tency of racial variation in medical outcomes among publicly and privately insured living kidney donors. Transplantation 2014;97:316– 324. Doshi MD, Goggins MO, Li L, Garg AX: Medical Outcomes in African American Live Kidney Donors: A Matched Cohort Study. Am J Transplant 2013;13:111–118. Muzaale AD, Massie AB, Wang MC, Montgomery RA, McBride MA, Wainright JL, et al: Risk of end-stage renal disease following live kidney donation. JAMA 2014; 311: 579– 586. Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JL Jr, et al: The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 Report. JAMA 2003; 289: 2560–2572. James PA, Oparil S, Carter BL, Cushman WC, Dennison-Himmelfarb C, Handler J, et al: 2014 evidence-based guideline for the management of high blood pressure in adults: report from the panel members appointed to the Eighth Joint National Committee (JNC 8). JAMA 2014;311:507–520.
Lentine/Schnitzler/Garg/Xiao/Axelrod/ Tuttle-Newhall/Brennan/Segev
Downloaded by: Univ. of California Santa Barbara 198.143.33.33 - 8/29/2015 12:43:21 PM
References
Antihypertensive Medication Use in Living Kidney Donors
21 Roy L, White-Guay B, Dorais M, Dragomir A, Lessard M, Perreault S: Adherence to antihypertensive agents improves risk reduction of end-stage renal disease. Kidney Int 2013; 84: 570–577. 22 Ibrahim HN, Foley R, Tan L, Rogers T, Bailey RF, Guo H, et al: Long-term consequences of kidney donation. N Engl J Med 2009; 360: 459–469. 23 Nogueira JM, Weir MR, Jacobs S, Haririan A, Breault D, Klassen D, et al: A study of renal outcomes in African-American living kidney donors. Transplantation 2009;88:1371–1376. 24 Vasan RS, Beiser A, Seshadri S, Larson MG, Kannel WB, D’Agostino RB, et al: Residual lifetime risk for developing hypertension in middle-aged women and men: The Framingham Heart Study. JAMA 2002;287:1003–1010. 25 Fehrman-Ekholm I, Elinder CG, Stenbeck M, Tyden G, Groth CG: Kidney donors live longer. Transplantation 1997;64:976–978. 26 Mjoen G, Reisaeter A, Hallan S, et al: Overall and cardiovascular mortality in Norwegian kidney donors compared to the background population. Nephrol Dial Transplant 2012; 27:443–447. 27 Gibney EM, Doshi MD, Hartmann EL, Parikh CR, Garg AX: Health insurance status of US living kidney donors. Clin J Am Soc Nephrol 2010;5:912–916.
28 Levy AR, O’Brien BJ, Sellors C, Grootendorst P, Willison D: Coding accuracy of administrative drug claims in the Ontario Drug Benefit database. Can J Clin Pharmacol 2003;10:67–71. 29 Lau HS, de Boer A, Beuning KS, Porsius A: Validation of pharmacy records in drug exposure assessment. J Clin Epidemiol 1997; 50: 619–625. 30 Boethius G: Recording of drug prescriptions in the county of Jamtland, Sweden. II. Drug exposure of pregnant women in relation to course and outcome of pregnancy. Eur J Clin Pharmacol 1977;12:37–43. 31 Leichtman A, Abecassis M, Barr M, Charlton M, Cohen D, Confer D, et al: Living kidney donor follow-up: state-of-the-art and future directions, conference summary and recommendations. Am J Transplant 2011;11:2561–2568. 32 Gaston RS, Young CJ: Living donor nephrectomy: understanding long-term risk in minority populations. Am J Transplant 2010;10: 2574–2576. 33 Lentine KL, Segev DL: Better understanding live donor risk through big data. Clin J Am Soc Nephrol 2013;8:1645.
Am J Nephrol 2014;40:174–183 DOI: 10.1159/000365157
183
Downloaded by: Univ. of California Santa Barbara 198.143.33.33 - 8/29/2015 12:43:21 PM
16 Lentine KL, Anyaegbu E, Gleisner A, Schnitzler MA, Axelrod D, Brennan DC, et al: Understanding medical care of transplant recipients through integrated registry and pharmacy claims data. Am J Nephrol 2013; 38: 420–429. 17 Buchanan PM, Schnitzler MA, Brennan DC, Dzebisashvili N, Willoughby LM, Axelrod D, et al: Novel methods for tracking longterm maintenance immunosuppression regimens. Clin J Am Soc Nephrol 2008; 3: 117– 124. 18 Lentine KL, Villines TC, Xiao H, Schnitzler MA, Brennan DC, Abbott KC, et al: Cardioprotective medication use after acute myocardial infarction in kidney transplant recipients. Transplantation 2011;91:1120–1126. 19 Pinsky BW, Takemoto SK, Lentine KL, Burroughs TE, Schnitzler MA, Salvalaggio PR: Transplant outcomes and economic costs associated with patient noncompliance to immunosuppression. Am J Transplant 2009; 9: 2597–2606. 20 Osterberg L, Blaschke T. Adherence to medication. N Engl J Med 2005;353:487–497.
