Approaches for dealing with missing data in health care studies

Penny, Kay I; Atkinson, Ian

doi:10.1111/j.1365-2702.2011.03854.x

Cited by 38 publications

(42 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although the extent of missing data for most items is below 5% (Table 1), restricting the analysis to clients having a response for all variables included in the model (a complete case analysis) reduces the sample from 5,144 to 3,734, a reduction of 27%. While a complete case approach is often used in health research [49], other methods are receiving increased attention with the choice of method depending on the pattern of missing data and the mechanisms causing it [50]. We believe that our missing data reflect a random process rather than systematic bias.…”

Section: Methodsmentioning

confidence: 99%

“…We believe that our missing data reflect a random process rather than systematic bias. However, we cannot be certain which of the three randomness patterns described in the literature applies to our data: missing completely at random (MCAR), missing at random (MAR), or missing not at random (MNAR) [50]. There is no universal method of handling MNAR, but the pattern is rare [51].…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Prevalence and risk factors of depressive symptoms in a Canadian palliative home care population: a cross-sectional study

Fisher

Seow

Brazil³

et al. 2014

BMC Palliat Care

View full text Add to dashboard Cite

BackgroundDepression in palliative care patients is important because of its intrinsic burden and association with elevated physical symptoms, reduced immunity and increased mortality risk. Identifying risk factors associated with depression can enable clinicians to more readily diagnose it, which is important since depression is treatable. The purpose of this cross-sectional study was to determine the prevalence of depressive symptoms and risk factors associated with them in a large sample of palliative home care patients.MethodsThe data come from interRAI Palliative Care assessments completed between 2006 and 2012. The sample (n = 5144) consists of adults residing in Ontario (Canada), receiving home care services, classified as palliative, and not experiencing significant cognitive impairment. Logistic regression identified the risk factors associated with depressive symptoms. The dependent variable was the Depression Rating Scale (DRS) and the independent variables were functional indicators from the interRAI assessment and other variables identified in the literature. We examined the results of the complete case and multiple imputation analyses, and found them to be similar.ResultsThe prevalence of depressive symptoms was 9.8%. The risk factors associated with depressive symptoms were (pooled estimates, multiple imputation): low life satisfaction (OR = 3.01 [CI = 2.37-3.82]), severe and moderate sleep disorders (2.56 [2.05-3.19] and 1.56 [1.18-2.06]), health instability (2.12 [1.42-3.18]), caregiver distress 2.01 [1.62-2.51]), daily pain (1.73 [1.35-2.22]), cognitive impairment (1.45 [1.13-1.87]), being female (1.37 [1.11-1.68]), and gastrointestinal symptoms (1.27 [1.03-1.55]). Life satisfaction mediated the effect of prognostic awareness on depressive symptoms.ConclusionsThe prevalence of depressive symptoms in our study was close to the median of 10-20% reported in the palliative care literature, suggesting they are present but by no means inevitable in palliative patients. Most of the factors associated with depressive symptoms in our study are amenable to clinical intervention and often targeted in palliative care programs. Designing interventions to address them can be challenging, however, requiring careful attention to patient preferences, the spectrum of comorbid conditions they face, and their social supports. Life satisfaction was one of the strongest factors associated with depressive symptoms in our study, and is likely to be among the most challenging to address.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Prevalence and risk factors of depressive symptoms in a Canadian palliative home care population: a cross-sectional study

Fisher

Seow

Brazil³

et al. 2014

BMC Palliat Care

View full text Add to dashboard Cite

show abstract

“…Given the number of respondents and the extensive number of questions, it is not surprising that non-core items included in the present analyses generated missing responses. Complete-case analyses were used given limited gain from undertaking complex missing-data imputation in such cases (Penny and Atkinson, 2012). Valid N is reported for each statistic.…”

Section: Discussionmentioning

confidence: 99%

Energy drink use frequency among an international sample of people who use drugs: Associations with other substance use and well-being

Peacock

Bruno

Ferris

et al. 2017

Drug and Alcohol Dependence

View full text Add to dashboard Cite

Highlights 74,864 people who use drugs participated in the online Global Drug Survey 2014  Energy drink, caffeine tablet and spray use was reported by 69.2%, 24.5% and 4.9%  Greater energy drink use frequency was associated with drug use  Greater use frequency was associated with hazardous drinking and poorer wellbeing 2 AbstractObjective: The study aims were to identify: i.) energy drink (ED), caffeine tablet, and caffeine intranasal spray use amongst a sample who report drug use, and ii.) the association between ED use frequency and demographic profile, drug use, hazardous drinking, and wellbeing.Method: Participants (n=74,864) who reported drug use completed the online 2014 Global Drug Survey. They provided data on demographics, ED use, and alcohol and drug use, completed the Alcohol Use Disorders Identification Test (AUDIT) and Personal Wellbeing Index (PWI), and reported whether they wished to reduce alcohol use.Results: Lifetime ED, caffeine tablet and intranasal caffeine spray use was reported by 69.2%, 24.5% and 4.9%. Median age of ED initiation was 16 years. For those aged 16 to 37, median years using EDs increased from 4 to 17 years of consumption, where it declined thereafter. Greater ED use frequency was associated with: being male; under 21 years of age; studying; and past year caffeine tablet/intranasal spray, tobacco, cannabis, amphetamine, MDMA, and cocaine use. Past year, infrequent (1-4 days) and frequent (≥5 days) past month ED consumers reported higher AUDIT scores and lower PWI scores than lifetime abstainers; past month consumers were less likely to report a desire to reduce alcohol use.Conclusions: ED use is part of a complex interplay of drug use, alcohol problems, and poorer personal wellbeing, and ED use frequency may be a flag for current/future problems.Prospective research is required exploring where ED use fits within the trajectory of other alcohol and drug use.

show abstract

“…Imputed means involves replacing missing data with a value (mean score) derived for an individual item from across the entire data set (Shrive et al, 2006). This process results in a complete data set being available for analysis (Penny & Atkinson, 2012), and has been shown to produce favourable results when used in scale development (Shrive et al, 2006). Hair et al (2010) suggest a four stage process to identify and select a suitable approach to dealing with missing data.…”

Section: Missing Datamentioning

confidence: 99%

“…The use of imputed means with MNAR data is problematic (Penny & Atkinson, 2012) and can lead to incorrect estimates (Eekhout et al, 2012). Therefore the data collected from all four surveys were used in the analysis, but imputed means was not used to replace missing data in these surveys.…”

Section: Missing Datamentioning

confidence: 99%

Moral distress and the Australian aged care worker: A mixed methods study

Burston¹

View full text Add to dashboard Cite

show abstract

Approaches for dealing with missing data in health care studies

Abstract: If nursing and healthcare practice is to be informed by research findings, then these findings must be reliable and valid. Researchers should report the details of missing data, and appropriate methods for dealing with missing values should be incorporated into the data analysis.

Cited by 38 publications

References 20 publications

Prevalence and risk factors of depressive symptoms in a Canadian palliative home care population: a cross-sectional study

Prevalence and risk factors of depressive symptoms in a Canadian palliative home care population: a cross-sectional study

Energy drink use frequency among an international sample of people who use drugs: Associations with other substance use and well-being

Moral distress and the Australian aged care worker: A mixed methods study

Contact Info

Product

Resources

About