Randomly and Non-Randomly Missing Renal Function Data in the Strong Heart Study: A Comparison of Imputation Methods

Toward a standardized evaluation of imputation methodology

Developing new imputation methodology has become a very active field. Unfortunately, there is no consensus on how to perform simulation studies to evaluate the properties of imputation methods. In part, this may be due to different aims between fields and studies. For example, when evaluating imputation techniques aimed at prediction, different aims may be formulated than when statistical inference is of interest. The lack of consensus may also stem from different personal preferences or scientific backgrounds. All in all, the lack of common ground in evaluating imputation methodology may lead to suboptimal use in practice. In this paper, we propose a move toward a standardized evaluation of imputation methodology. To demonstrate the need for standardization, we highlight a set of possible pitfalls that bring forth a chain of potential problems in the objective assessment of the performance of imputation routines. Additionally, we suggest a course of action for simulating and evaluating missing data problems. Our suggested course of action is by no means meant to serve as a complete cookbook, but rather meant to incite critical thinking and a move to objective and fair evaluations of imputation methodology. We invite the readers of this paper to contribute to the suggested course of action.

One night of sleep fragmentation does not affect exercise-induced leukocyte trafficking or mitogen-stimulated leukocyte oxidative burst in healthy men

PURPOSE

This study examined whether one night of sleep fragmentation alters circulating leukocyte counts and mitogen-stimulated oxidative burst by leukocytes at rest and in response to an acute bout of vigorous exercise.

METHODS

In a randomised crossover design, nine healthy men (mean ± SD: age 22 ± 2 years; BMI 24.9 ± 1.9 kg/m²) were exposed to one night of fragmented or uninterrupted sleep before cycling for 45 min at 71% ± 4% V̇O_2peak. Finger-tip blood samples were collected at rest, immediately post-exercise and one-hour post-exercise. Total leukocytes, lymphocytes, monocytes and neutrophils were counted. Leukocyte oxidative burst was assessed in whole blood by measuring Reactive Oxygen Species (ROS) production with luminol-amplified chemiluminescence after stimulation with phorbol 12-myristate 13-acetate (PMA).

RESULTS

Exercise elicited the expected trafficking pattern of leukocytes, lymphocytes, monocytes and neutrophils. Compared to rest, PMA-stimulated ROS production was increased one-hour post-exercise (+73% ± 65%; p = 0.019; data combined for fragmented and uninterrupted sleep). There were no statistically significant effects of fragmented sleep on leukocyte, lymphocyte, monocyte, and neutrophil counts or on ROS production at rest, immediately post-exercise or one-hour post-exercise (p > 0.05). However, with fragmented sleep, there was a +10% greater lymphocytosis immediately post-exercise (fragmented +40% ± 37%; uninterrupted +30% ± 35%; p = 0.51) and a -19% smaller neutrophilia by one-hour post-exercise (fragmented +103% ± 88%; uninterrupted +122% ± 131%; p = 0.72).

CONCLUSION

Fragmented sleep did not substantially alter the magnitude or pattern of exercise-induced leukocyte trafficking or mitogen-stimulated oxidative burst by leukocytes.

Design and respondent selection of a population-based study on associations between breast cancer screening, lifestyle and quality of life

BACKGROUND

Only few studies have integrated breast cancer screening, lifestyle, and quality of life. Potential bias due to selective non-response may disrupt associations being investigated. We describe the design of a Finnish population-based study on associations between breast cancer screening and various indicators for lifestyle and quality of life, and evaluate the level of bias among the respondents from the first study rounds over 2 years.

METHODS

The study target population of 10,000, 49-year-old women was randomly drawn from the Finnish National Population Registry. The data included birth year, marital status, municipality, and primary language. Data on education were retrieved from Statistics Finland. Questionnaires focusing on lifestyle-related risk factors and quality of life were sent to the target population in 2012-13, 1 year before the first invitation to organized breast cancer screening. We evaluated associations between willingness to respond and demographic characteristics in the eligible study population. Additionally, we examined associations between the demographic characteristics and the Satisfaction With Life Scale (SWLS), and evaluated the impact of non-response using inverse probability weighting and multiple imputation.

RESULTS

The questionnaire response proportion was 52.4%. Compared to non-respondents, respondents were more often married, academically educated, and native speakers of Finnish or Swedish. Nevertheless, the estimates of the SWLS among the respondents were in line with those corrected by non-response in the eligible study population.

CONCLUSIONS

Based on the SWLS, the respondents are representative of women in the entire eligible study population.