Statistical modelling for falls count data
- Authors: Ullah, Shahid , Finch, Caroline , Day, Lesley
- Date: 2010
- Type: Text , Journal article
- Relation: Accident Analysis and Prevention Vol. 42, no. 2 (2010), p. 384-392
- Relation: http://purl.org/au-research/grants/nhmrc/565900
- Full Text:
- Reviewed:
- Description: Falls and their injury outcomes have count distributions that are highly skewed toward the right with clumping at zero, posing analytical challenges. Different modelling approaches have been used in the published literature to describe falls count distributions, often without consideration of the underlying statistical and modelling assumptions. This paper compares the use of modified Poisson and negative binomial (NB) models as alternatives to Poisson (P) regression, for the analysis of fall outcome counts. Four different count-based regression models (P, NB, zero-inflated Poisson (ZIP), zero-inflated negative binomial (ZINB)) were each individually fitted to four separate fall count datasets from Australia, New Zealand and United States. The finite mixtures of P and NB regression models were also compared to the standard NB model. Both analytical (F, Vuong and bootstrap tests) and graphical approaches were used to select and compare models. Simulation studies assessed the size and power of each model fit. This study confirms that falls count distributions are over-dispersed, but not dispersed due to excess zero counts or heterogeneous population. Accordingly, the P model generally provided the poorest fit to all datasets. The fit improved significantly with NB and both zero-inflated models. The fit was also improved with the NB model, compared to finite mixtures of both P and NB regression models. Although there was little difference in fit between NB and ZINB models, in the interests of parsimony it is recommended that future studies involving modelling of falls count data routinely use the NB models in preference to the P or ZINB or finite mixture distribution. The fact that these conclusions apply across four separate datasets from four different samples of older people participating in studies of different methodology, adds strength to this general guiding principle. © 2009 Elsevier Ltd. All rights reserved.
Linked versus unlinked hospital discharge data on hip fractures for estimating incidence and comorbidity profiles
- Authors: Vu, Trang , Day, Lesley , Finch, Caroline
- Date: 2012
- Type: Text , Journal article
- Relation: BMC Medical Research Methodology Vol. 12, no. 113 (2012), p. 1-8
- Relation: http://purl.org/au-research/grants/nhmrc/565900
- Full Text:
- Reviewed:
- Description: Background: Studies comparing internally linked (person–identifying) and unlinked (episodes of care) hospital discharge data (HDD) on hip fractures have mainly focused on incidence overestimation by unlinked HDD, but little is known about the impact of overestimation on patient profiles such as comorbidity estimates. In view of the continuing use of unlinked HDD in hip fracture research and the desire to apply research results to hip fracture prevention, we concurrently assessed the accuracy of both incidence and comorbidity estimates derived from unlinked HDD compared to those estimated from internally linked HDD. Methods: We analysed unlinked and internally linked HDD between 01 July 2005 and 30 June 2008, inclusive, from Victoria, Australia to estimate the incidence of hospital admission for fall-related hip fracture in community-dwelling older people aged 65+ years and determine the prevalence of comorbidity in patients. Community-dwelling status was defined as living in private residence, supported residential facilities or special accommodation but not in nursing homes. We defined internally linked HDD as the reference standard and calculated measures of accuracy of fall-related hip fracture incidence by unlinked HDD using standard definitions. The extent to which comorbidity prevalence estimates by unlinked HDD differed from those by the reference standard was assessed in absolute terms. Results: The sensitivity and specificity of a standard approach for estimating fall-related hip fracture incidence using unlinked HDD (i.e. omitting records of in-hospital deaths, inter-hospital transfers and readmissions within 30 days of discharge) were 94.4% and 97.5%, respectively. The standard approach and its variants underestimated the prevalence of some comorbidities and altered their ranking. The use of more stringent selection criteria led to major improvements in all measures of accuracy as well as overall and specific comorbidity estimates. Conclusions: This study strongly supports the use of linked rather than unlinked HDD in injury research. In health systems where linked HDD are unavailable, current approaches for identifying incident hip fractures may be enhanced by incorporating additional evidence-based criteria.
Accuracy of evidence-based criteria for identifying an incident hip fracture in the absence of the date of injury: a retrospective database study
- Authors: Vu, Trang , Davie, Gabrielle , Barson, David , Day, Lesley , Finch, Caroline
- Date: 2013
- Type: Text , Journal article
- Relation: BMJ Open Vol. 3, no. 7 (2013), p. 1-6
- Relation: http://purl.org/au-research/grants/nhmrc/565900
- Full Text:
- Reviewed:
- Description: Objectives: Hospital discharge data (HDD) in many health systems do not capture the date of injury (DOI); the absence of this date hinders researchers’ ability to distinguish repeat from incident injury admissions. Various approaches using somewhat arbitrary criteria have been explored to increase the accuracy of incident injury identification. However, these approaches have not been validated against a data source which contains DOI. The aim of this study was to determine the accuracy of evidence-based criteria for identifying fall-related incident hip fractures in the absence of DOI using HDD containing DOI as the reference standard. Design: Retrospective database study. Setting: New Zealand. Participants: 8761 patients aged 65+ years admitted for fall-related hip fracture between 1 July 2005 and 30 June 2008, inclusive. Outcome measures: We defined person-identifying HDD containing DOI as the reference standard and calculated measures of the accuracy of evidence-based criteria for identifying fall-related incident hip fractures from HDD not containing DOI. The criteria were principal diagnosis of hip fracture, mechanism of injury indicating a fall, admission type emergency, admission source other than a transfer and presence of hip operation code(s). For a subsequent fall-related hip fracture, additional criteria were time between successive hip fractures ≥120 days, and all external cause-of-injury codes being different to those for the previous hip fracture. Results: The sensitivity and specificity of the criteria for identifying fall-related incident hip fractures from data not containing DOI were 96.7% and 99.3%, respectively, compared with the reference standard. The application of these criteria resulted in a slight underestimation of the percentage of patients with multiple hip fractures. Conclusions: Although it is preferable to have DOI; this study demonstrates that evidence-based criteria can be used to reliably identify fall-related incident hip fractures from the person-identifying HDD when DOI is unavailable.
Patterns of comorbidity in community-dwelling older people hospitalised for fall-related injury: A cluster analysis
- Authors: Vu, Trang , Finch, Caroline , Day, Lesley
- Date: 2011
- Type: Text , Journal article
- Relation: BMC Geriatrics Vol. 11, no. 45 (2011), p. 1-10
- Relation: http://purl.org/au-research/grants/nhmrc/565900
- Full Text:
- Reviewed:
- Description: Background: Community-dwelling older people aged 65+ years sustain falls frequently; these can result in physical injuries necessitating medical attention including emergency department care and hospitalisation. Certain health conditions and impairments have been shown to contribute independently to the risk of falling or experiencing a fall injury, suggesting that individuals with these conditions or impairments should be the focus of falls prevention. Since older people commonly have multiple conditions/impairments, knowledge about which conditions/impairments coexist in at-risk individuals would be valuable in the implementation of a targeted prevention approach. The objective of this study was therefore to examine the prevalence and patterns of comorbidity in this population group. Methods: We analysed hospitalisation data from Victoria, Australia's second most populous state, to estimate the prevalence of comorbidity in patients hospitalised at least once between 2005-6 and 2007-8 for treatment of acute fall-related injuries. In patients with two or more comorbid conditions (multicomorbidity) we used an agglomerative hierarchical clustering method to cluster comorbidity variables and identify constellations of conditions. Results: More than one in four patients had at least one comorbid condition and among patients with comorbidity one in three had multicomorbidity (range 2-7). The prevalence of comorbidity varied by gender, age group, ethnicity and injury type; it was also associated with a significant increase in the average cumulative length of stay per patient. The cluster analysis identified five distinct, biologically plausible clusters of comorbidity: cardiopulmonary/metabolic, neurological, sensory, stroke and cancer. The cardiopulmonary/metabolic cluster was the largest cluster among the clusters identified. Conclusions: The consequences of comorbidity clustering in terms of falls and/or injury outcomes of hospitalised patients should be investigated by future studies. Our findings have particular relevance for falls prevention strategies, clinical practice and planning of follow-up services for these patients.
A protocol for evidence-based targeting and evaluation of statewide strategies for preventing falls among community-dwelling older people in Victoria, Australia
- Authors: Day, Lesley , Finch, Caroline , Hill, Keith , Haines, Terry , Clemson, Lindy , Thomas, Margaret , Thompson, Catherine
- Date: 2011
- Type: Text , Journal article
- Relation: Injury Prevention Vol. 17, no. 2 (2011), p. 1-8
- Relation: http://purl.org/au-research/grants/nhmrc/565900
- Relation: http://purl.org/au-research/grants/nhmrc/546282
- Full Text:
- Reviewed:
- Description: Background: Falls are a significant threat to the safety, health and independence of older citizens. Despite the now substantial evidence about effective falls prevention interventions, translation into falls reductions has not yet been fully realised. While the hip fracture rate is decreasing, the number and rate of fall-related hospital admissions among older people is increasing. The challenge now is to deliver the most effective interventions efficiently at a population level, and for these interventions to be taken up by older people. Objective: To support the development, and evaluation of, effective falls prevention policy and practice in the state of Victoria, Australia. Methods: The RE-AIM model (Reach, Efficacy, Adoption, Implementation, Maintenance) was used to identify strategies for an effective programme. Research objectives were developed to support the strategies. These include: (1) identification of subgroups of older people most frequently admitted to hospital for falls; (2) examining the acceptability of established falls interventions; (3) identification of factors that encourage and support relevant lifestyle changes; (4) identifying opportunities to incorporate confirmed interventions in existing programmes and services; (5) developing guidelines for sustainability. The research results will subsequently guide strategy details for the falls prevention plan. RE-AIM will provide the framework for the evaluation structure. Outcome measures: Measures to monitor the implementation of the selected interventions will be determined for each intervention, based on the five key factors of the RE-AIM model. The overall effect of the falls prevention plan will be monitored by time series analysis of fall-related hospital admission rates for community-dwelling older people.
Investigation of older adults’ participation in exercises following completion of a state-wide survey targeting evidence-based falls prevention strategies
- Authors: Lee, Den-Ching , Day, Lesley , Finch, Caroline , Hill, Keith , Clemson, Lindy , McDermott, Fiona , Haines, Terry
- Date: 2014
- Type: Text , Journal article
- Relation: Journal of Aging and Physical Activity Vol.23, no.2 (2014), p.256-263
- Relation: http://purl.org/au-research/grants/nhmrc/565900
- Full Text:
- Reviewed:
- Description: This paper examines whether involvement in an observational study may prompt participants to change their exercise behaviors. Data was collected from 394 older community dwellers in Victoria, Australia using a baseline survey, and 245 of these participated in a follow-up survey one year later. Survey domains were drawn from constructs of relevant health behavior models. Results showed that the proportion of respondents who were currently participating in exercises to prevent falls at follow-up was 12% higher than at baseline (Wilcoxon p value<0.001). Twenty-nine percent reported they had changed their perceptions about falls and their risk of falls, with comments focused on threat appraisal. Forty-four percent reported having taken strategies to reduce their risk of falling, with comments based on implementation of different preventive strategies. Respondents who held favorable views towards exercises for the prevention of falls appear to change their behaviors that might address falls when participating in observational studies.