Sampling and external validity

Utilisation of outcome measures in the management of non-specific neck pain: A national survey of current physiotherapy practice in the UK

BACKGROUND

Routine utilisation of outcome measures (OMs) is an integral part of physiotherapy rehabilitation when managing non-specific neck pain (NSNP). Numerous relevant OMs exist; however, the extent to which OMs are used by physiotherapists in the UK for NSNP is unknown.

OBJECTIVE

To determine current utilisation patterns of OMs in UK physiotherapy practice when managing NSNP.

METHODS

An online web-based survey instrument was developed and physiotherapists were invited to participate if they were currently practicing in the UK and had some experience of managing patients with neck pain. Logistic regression analyses using the generalised linear models was carried out to identify characteristics associated with OMs utilisation.

RESULTS

A total of 2101 surveys were completed. One-third of the respondents reported not using OMs when managing NSNP. Lack of time and clear guidance about the suitability of available OMs were the most commonly reported reasons. A majority of the two-thirds of those who reported using OMs were consistently using pain and range of motion rating measures. Physical/functional limitations, psychological distress, and quality of life constructs, which are frequently associated with NSNP, were rarely measured. Years of practice (p = 0.000), nation (p = 0.019) and proportion of patients with neck pain (p = 0.034) variables were found to be independently associated with frequent use of OMs.

CONCLUSION

This survey established the poor integration of OMs in the UK when managing NSNP. Further attention is required to identify or develop OMs which are feasible for use in busy clinical practice and to market them more effectively to physiotherapists.

Exploring the role of GIS during community health assessment problem solving: experiences of public health professionals

BACKGROUND

A Community health assessment (CHA) involves the use of Geographic Information Systems (GIS) in conjunction with other software to analyze health and population data and perform numerical-spatial problem solving. There has been little research on identifying how public health professionals integrate this software during typical problem solving scenarios. A better understanding of this is needed to answer the "What" and the "How". The "What" identifies the specific software being used and the "How" explains the way they are integrated together during problem solving steps. This level of understanding will highlight the role of GIS utilization during problem solving and suggest to developers how GIS can be enhanced to better support data analysis during community health assessment.

RESULTS

An online survey was developed to identify the information technology used during CHA analysis. The tasks were broken down into steps and for our analysis these steps were categorized by action: Data Management/Access, Data Navigation, Geographic Comparison, Detection of Spatial Boundaries, Spatial Modelling, and Ranking Analysis. 27 CHA professionals completed the survey, with the majority of participants (14) being from health departments. Statistical software (e.g. SPSS) was the most popular software for all but one of the types of steps. For this step (detection of spatial boundaries), GIS was identified as the most popular technology.

CONCLUSION

Most CHA professionals indicated they use statistical software in conjunction with GIS. The statistical software appears to drive the analysis, while GIS is used primarily for simple spatial display (and not complex spatial analysis). This purpose of this survey was to thoroughly examine into the process of problem solving during community health assessment data analysis and to gauge how GIS is integrated with other software for this purpose. These findings suggest that GIS is used more for spatial display while other software such as statistical packages (the "What") are used to drive data management, data navigation, and data calculation (the "How"). Focusing at the level of how public health problems are solved, can shed light on how GIS technology can be enhanced to encompass a stronger role during community health assessment problem solving.

Predicting girls' conduct disorder in adolescence from childhood trajectories of disruptive behaviors

OBJECTIVE

To examine girls' developmental trajectories of disruptive behaviors during the elementary school years and to predict conduct disorder symptoms and diagnosis in adolescence with trajectories of these behaviors.

METHOD

The sample was 820 girls from the province of Quebec followed over 10 years (1986-1996). A semiparametric mixture model was used to describe girls' developmental trajectories of teacher-rated disruptive behaviors between the ages of 6 and 12 years. The trajectories were used to predict conduct disorder symptoms and diagnosis when the girls were on average 15.7 years.

RESULTS

Four groups of girls following trajectories with distinct levels of disruptive behaviors were identified: a low, medium, medium-high, and high trajectory. Prediction with the trajectories indicated that girls on the medium, medium-high, and high trajectories reported a significantly higher number of conduct disorder symptoms in adolescence. However, only the girls on the medium-high and high trajectories were at significantly higher risk to meet DSM-III-R criteria for conduct disorder, compared with girls in the low group (odds ratio: 4.46). More than two thirds of the girls with conduct disorder were in the medium or higher-level trajectories.

CONCLUSION

The results suggest that there is an early-onset type of conduct disorder in girls.