A Review of Hot Deck Imputation for Survey Non-response

Content and timing of antenatal care: predisposing, enabling and pregnancy-related determinants of antenatal care trajectories

BACKGROUND

When examining risk factors for inadequate antenatal care, the assessment of antenatal care hardly considers the content and timing of interventions during pregnancy. This study aims to provide information about the importance of predisposing, enabling and pregnancy-related determinants on the received content and timing of antenatal care.

METHODS

In the Brussels Metropolitan Region, 333 women were consecutively recruited at the beginning of their pregnancies. Antenatal care use was recorded prospectively. A classification system measuring the content and timing of care during pregnancy (CTP) divided the women into four categories. Ordinal regression analyses were applied to define unadjusted and adjusted odds ratios (ORs), measuring the effect of different determinants on being assigned to a higher CTP category.

RESULTS

A total of 10.2% of the women had an inadequate, 8.4% an intermediate, 36% a sufficient and 45.3% an appropriate antenatal care trajectory. Adjusted ORs showed a lower likelihood of being assigned to a higher CTP category for lower educated women (OR: 0.58; 95% CI 0.37-0.92), women of Maghreb origin (OR: 0.38; 95% CI 0.22-0.66) and women with a higher discontinuity of care (OR: 0.56; 95% CI 0.34-0.90).

CONCLUSIONS

When controlling for confounders, no enabling determinants were found that affect the content and timing of care in pregnancy. Although antenatal care is equally available to all women, predisposing and pregnancy-related factors were related to the likelihood of completing an appropriate antenatal care trajectory. Besides stimulating knowledge about the importance of antenatal care in less-educated and Maghreb women, supporting continuity of care during pregnancy might result in higher levels of received antenatal care.

The influence of age, gender and socio-economic status on multimorbidity patterns in primary care. First results from the multicare cohort study

BACKGROUND

Multimorbidity is a phenomenon with high burden and high prevalence in the elderly. Our previous research has shown that multimorbidity can be divided into the multimorbidity patterns of 1) anxiety, depression, somatoform disorders (ADS) and pain, and 2) cardiovascular and metabolic disorders. However, it is not yet known, how these patterns are influenced by patient characteristics. The objective of this paper is to analyze the association of socio-demographic variables, and especially socio-economic status with multimorbidity in general and with each multimorbidity pattern.

METHODS

The MultiCare Cohort Study is a multicentre, prospective, observational cohort study of 3.189 multimorbid patients aged 65+ randomly selected from 158 GP practices. Data were collected in GP interviews and comprehensive patient interviews. Missing values have been imputed by hot deck imputation based on Gower distance in morbidity and other variables. The association of patient characteristics with the number of chronic conditions is analysed by multilevel mixed-effects linear regression analyses.

RESULTS

Multimorbidity in general is associated with age (+0.07 chronic conditions per year), gender (-0.27 conditions for female), education (-0.26 conditions for medium and -0.29 conditions for high level vs. low level) and income (-0.27 conditions per logarithmic unit). The pattern of cardiovascular and metabolic disorders shows comparable associations with a higher coefficient for gender (-1.29 conditions for female), while multimorbidity within the pattern of ADS and pain correlates with gender (+0.79 conditions for female), but not with age or socioeconomic status.

CONCLUSIONS

Our study confirms that the morbidity load of multimorbid patients is associated with age, gender and the socioeconomic status of the patients, but there were no effects of living arrangements and marital status. We could also show that the influence of patient characteristics is dependent on the multimorbidity pattern concerned, i.e. there seem to be at least two types of elderly multimorbid patients. First, there are patients with mainly cardiovascular and metabolic disorders, who are more often male, have an older age and a lower socio-economic status. Second, there are patients mainly with ADS and pain-related morbidity, who are more often female and equally distributed across age and socio-economic groups.

TRIAL REGISTRATION

ISRCTN89818205.

Predisposing, enabling and pregnancy-related determinants of late initiation of prenatal care

Prenatal care is important for the health and wellbeing of women and their babies. There is international consensus that prenatal care should begin in the first trimester. This study aims to analyze the effects of predisposing, enabling and pregnancy-related determinants of late prenatal care initiation. In this prospective observational study, 333 women were recruited consecutively at the beginning of their prenatal care trajectory. Data was collected on the timing of the first prenatal visit and on socio-demographic and pregnancy-related characteristics, using a semi-structured interview. A multivariate binominal logistic regression was applied to analyze independent effects on late initiation of prenatal care. Bivariately late initiation of care was associated with being inactive on the labor market, non-European origin, not having lived in Belgium since birth, low income, receiving welfare benefits, not having a regular obstetrician and experiencing difficulties getting a first appointment. When adjusting for all determinants, our multivariate analyses showed that late initiation was associated with non-European origin, low income and not having a regular obstetrician. This study shows that late initiation of prenatal care is associated with predisposing and enabling determinants. In order to ensure timely initiation of care, policy-makers should focus on encouraging women to have a regular prenatal care provider before pregnancy and taking steps in lowering out-of-pocket fees for low-income women. Future research is needed to examine whether these determinants are associated with initiation of care only or whether they play a role in the pregnancy follow-up as well.

A hot-deck multiple imputation procedure for gaps in longitudinal recurrent event histories

We propose a regression-based hot-deck multiple imputation method for gaps of missing data in longitudinal studies, where subjects experience a recurrent event process and a terminal event. Examples are repeated asthma episodes and death, or menstrual periods and menopause, as in our motivating application. Research interest concerns the onset time of a marker event, defined by the recurrent event process, or the duration from this marker event to the final event. Gaps in the recorded event history make it difficult to determine the onset time of the marker event, and hence, the duration from onset to the final event. Simple approaches such as jumping gap times or dropping cases with gaps have obvious limitations. We propose a procedure for imputing information in the gaps by substituting information in the gap from a matched individual with a completely recorded history in the corresponding interval. Predictive mean matching is used to incorporate information on longitudinal characteristics of the repeated process and the final event time. Multiple imputation is used to propagate imputation uncertainty. The procedure is applied to an important data set for assessing the timing and duration of the menopausal transition. The performance of the proposed method is assessed by a simulation study.

Handling missing data issues in clinical trials for rheumatic diseases

Missing data are ubiquitous in clinical trials for rheumatic diseases, and it is important to accommodate them using appropriate statistical techniques. We review some of the basic considerations for missing data and survey a range of statistical techniques for analysis of longitudinal clinical trial data with missingness. Using clinical trial data from patients with diffuse systemic sclerosis, we show that different approaches to handling missing data can lead to different conclusions on the efficacy of the treatment. We then suggest how such discrepancies might be addressed. In particular, we emphasize that the commonly used method in rheumatic clinical trials of carrying the last observation forward to impute missing values should not be the primary analysis. We review software for analyzing different types of missing data and discuss our freely available software library for analyzing the more difficult but more realistic situation when the probability of dropout or missing data may depend on the unobserved missing value.

The Use of Sample Weights in Hot Deck Imputation

A common strategy for handling item nonresponse in survey sampling is hot deck imputation, where each missing value is replaced with an observed response from a "similar" unit. We discuss here the use of sampling weights in the hot deck. The naive approach is to ignore sample weights in creation of adjustment cells, which effectively imputes the unweighted sample distribution of respondents in an adjustment cell, potentially causing bias. Alternative approaches have been proposed that use weights in the imputation by incorporating them into the probabilities of selection for each donor. We show by simulation that these weighted hot decks do not correct for bias when the outcome is related to the sampling weight and the response propensity. The correct approach is to use the sampling weight as a stratifying variable alongside additional adjustment variables when forming adjustment cells.