This study has a number of strengths. It is a multi-site study of a population of Veterans that leverages detailed laboratory and electronic health record data to establish rates of test measurement among a cohort with a high prevalence of CKD, diabetes, and hypertension. The study adds to the literature by providing estimates of surveillance among both the general high risk cohort of AKI survivors as well as the sub-group with pre-existing CKD. This study also included sensitivity analyses and evaluation for more severe degrees of injury. Stratification by baseline eGFR did show some threshold effects, GW-572016 particularly when the baseline was close to 60 mL/min/1.73 m2. However this reflects a clinically significant threshold for care, so we included analyses for the whole population and stratified by baseline eGFR to improved interpretation flexibility. In addition, severity of injury was noted to impact the rapidity but not the final rate of recovery and receipt of PTH and phosphorus measurement. We recognize several study limitations. The veteran patient population may limit generalizability to other care settings. Excluding patients with lengths of stay greater than 30 days may potentially bias the population towards a healthier one, but this represents a very different sub-population within the VA, largely among patients extended rehabilitation, placement or psychosocial barriers to discharge, and for these reasons we excluded them. As noted above, we also did not evaluate the test result values in the follow-up period to determine whether they were abnormal or if clinical care changed in response to test ascertainment, and so a portion of those tested may not have received the recommended care following surveillance. While providers may have appropriately used ACEi or ARBs and optimally controlled BP without quantifying proteinuria, the literature on quality of CKD care suggests this is unlikely. In addition, it is possible that some patients obtaining care at an outside healthcare facility were not captured. However, restriction of the cohort to those patients with frequent contacts with the VA likely limited this bias towards not receiving the recommended care, and rates of serum creatinine measurement were quite high, indicating that laboratory surveillance was conducted on the majority of these patients within the VA. Lastly, as optimal post-AKI management remains to be defined, it is yet unknown the extent that modifying risk factors such as proteinuria improve outcomes in this patient population and will likely require formal testing in future trials.