Missing data is a common problem in clinical epidemiology research. Inappropriate handling of missing data leads to biased results. This paper explains the mechanisms of missing data and several methods for handling missing data. In particular, multiple imputation is a more valid approach than other methods. Therefore, this paper focuses on the assumptions and procedures for multiple imputation and describes its limitations.

In this era of large-scale, multi-institutional studies, the importance of analyzing hierarchical (clustered) data is increasing. However, conventional regression analysis may be inadequate for this purpose because it assumes that records for individual patients are independent of records for other patients. Multilevel analysis is a statistical method that allows one to analyze data with a hierarchical structure. Mixed-effect models expand the conventional regression models by incorporating random coefficients for each unit of cluster. Multilevel analysis can be applied to studies on repeated measures among individuals, multi-institutional studies on effects of individual-level variables, and studies on effects of cluster-level variables or cluster-level variances. This report summarizes the basics of multilevel analysis and how it can be used in clinical epidemiology research.

BACKGROUND

Survival benefit of outpatient cardiac rehabilitation (CR) after acute myocardial infarction (AMI) has recently been contested under the current real-world clinical practice. We investigated whether outpatient CR was associated with lower mortality and morbidity risks among Japanese AMI patients.

METHODS

We analyzed patients who were admitted for AMI and received both percutaneous coronary intervention and inpatient CR from January 2011 to December 2014, using a nationwide administrative database in Japan (final date of follow-up: July 31, 2016). We compared patients who received outpatient CR and who did not, and the primary outcome was a composite of all-cause death and recurrence of AMI after the landmark time-point of day 180 after discharge. We applied Cox proportional hazards model to estimate outcomes, and propensity-score matching was applied to adjust for baseline imbalances.

RESULTS

A total of 5,654 patients (mean [SD] age, 66.8 [12.4] years; 21.2% female; median follow-up period [IQR] 1.44 [0.87, 2.27] years), 730 (12.9%) participated in outpatient CR at least once within 180 days of discharge. Of 1,458 propensity-score matched patients, outpatient CR participation was associated with lower but statistically non-significant risks among the primary outcome (1.38 vs. 2.12 per 100 patient-years; HR = 0.71; 95%CI, 0.32 to 1.61).

CONCLUSIONS

Among Japanese patients who admitted for AMI and received both percutaneous coronary intervention and inpatient CR, outpatient CR was underutilized, and associated with a statistically non-significant mortality and morbidity benefits. Further study is necessary to reaffirm the real-world effectiveness of outpatient CR under the current real-world clinical practice.

BACKGROUND

Although anemia is an important factor determining renal and cardiovascular outcomes in patients with chronic kidney disease (CKD), reports on the status of anemia management in Japanese patients with pre-dialysis CKD are limited.

METHODS

We retrospectively analyzed CKD patients follow-up at our department between June 2012 and July 2014 (Fukushima CKD cohort study), for the management status of anemia and differences in anemia control under the current treatment strategy, including long-acting erythropoiesis stimulating agents (ESA), classified according to the underlying renal disease, namely primary renal disease, hypertensive nephropathy, diabetic nephropathy, and others. Anemia was defined as a hemoglobin level of <11 g/dL, or receiving ESA therapy.

RESULTS

We identified 1,324 patients with CKD, and found that the prevalence of anemia increased with CKD stage, and was significantly higher in diabetic nephropathy than primary renal disease and hypertensive nephropathy (25.3%, 12.9%, and 10.9%, respectively, P <0.001). Anemia was independently related to the underlying renal disease, with a significant difference between diabetic nephropathy and primary renal disease (odds ratio 2.15; 95% confidence interval, 1.16–3.99, P = 0.015). Among those with hemoglobin <11 g/dL, 57.1% of patients were not on ESAs. The mean hemoglobin level was 10.4 ± 1.0 g/dL and achievement rate of a target hemoglobin of ≥11 g/dL, recommended in Japanese guidelines, was 30.1% among subjects on ESA therapy.

CONCLUSIONS

These results suggest that intervention for renal anemia is not necessarily adequate among Japanese patients with CKD, even those under nephrologist care, including with use of long-acting ESAs.