Susan C. Scott

Statistical assessment of ordinal outcomes in comparative studies

Ordinal regression is a relatively new statistical method developed for analyzing ranked outcomes. In the past, ranked scales have often been analyzed without making full use of the ordinality of the data or, alternatively, by assigning arbitrary numerical scores to the ranks. While ordinal regression models are now available to make full use of ranked data, they are not used widely. This article, directed to clinical researchers and epidemiologists, provides a description of the properties of these methods. Using ordinal measures of back pain in a follow-up study of adolescent idiopathic scoliosis, we illustrate the advantages of those methods and describe how to interpret the estimated parameters. Comparisons with binary logistic regression are made to show why a single dichotomization of the ordinal scale may lead to incorrect inferences. Two ordinal models (the proportional odds and the continuation ratio models) are discussed, and the goodness-of-fit of these models is examined. We conclude that ordinal regression is a tool that is powerful, simple to use, and produces an interpretable parameter that summarizes the effect between groups over all levels of the outcome.

A review of the association between cigarette smoking and the development of nonspecific back pain and related outcomes.

Study Design. A structured review of the epidemiologic literature was performed. Thirty-eight studies published in peer-reviewed journals were reviewed. The methodologic strengths and weaknesses of the studies were described and assessed qualitatively. Four studies were excluded because of difficulties in design or interpretation. Objectives. To provide a systematic analysis of the literature to assess the evidence as to whether smoking is associated with the prevalence and incidence of nonspecific back pain and related outcomes. Summary of Background Data. Evidence has been gathering regarding the association of smoking with nonspecific back pain and other back disorders, but a comprehensive summary and evaluation of the data have not been published. Results. Positive associations between current smoking and nonspecific back pain were found in 18 of 26 studies in men and 18 of 20 studies in women. For sciatica and herniated discs, there were four of eight and one of five positive studies in men and women, respectively. The majority of these studies were cross-sectional (18 in men and 16 in women), with only a handful of prospective studies. Positive associations between past smoking and nonspecific back pain were reported in five of nine studies in men and five of six studies in women. In addition, increases in the prevalence and/or incidence of nonspecific back pain were found in the majority of studies in which level of consumption was analyzed and reported. An attempt was made to assess whether these results could be artifactual arising from selection bias, confounding bias, publication bias, or errors in measurement. As well, the biologic mechanisms were summarized that have been suggested by various investigators. Conclusions. The available data are consistent with the notion that smoking is associated with the incidenceand prevalence of nonspecific back pain, but there are too few studies to make any conclusions for the other end points (e.g., sciatica, herniated discs). It cannot be ruled out that the association is a statistical artifact arising from either selection or confounding factors, because the evidence for nonspecific low back pain derives mostly from cross-sectional studies. In addition, it cannot be stated unequivocally that smoking preceded back pain. Long-term follow-up studies are needed to eliminate the possibility that chronic back pain preceded smoking, to better estimate dose–response correlations, and to perform biologic measurements to elucidate possible mechanisms.

The prognostic consequences in the making of the initial medical diagnosis of work-related back injuries.

Study Design A cohort of 1848 workers, representative of all sectors of industry, who were compensated for a low back injury in 1988 but not in the previous 2 years, was followed over 24 months. Objectives To determine the prognostic value of the physicians initial diagnosis of back problems. Summary of Background Data In the absence of a standardized classification of diagnoses of back pain, this study aimed to provide an element of validity to a classification previously proposed that consists of “specific” and “nonspecific” back pain. Methods Medical charts were reviewed at the Quebec Workers Compensation Board to extract the diagnosis made by the treating physicians within 7 days of the first day of absence from work. Diagnoses were categorized into “specific” (lesions of vertebrae and discs) and “nonspecific” [pain, sprains, and strains). The history of compensated work absence for low back pain in the following 24 months was obtained. Results A specific diagnosis was found in 8.9% (165) of the workers, accounting for 31.0% of the patients who accumulated 6 months or more of absence in 2 years. Increasing age and daily amount of compensation also were associated with an increased risk of chronicity. Conclusions The physicians initial diagnosis was highly associated with the risk of chronicity. The explanation for this result is complex, involving the nature of the underlying lesion as well as the impact of the diagnosis “label” on the worker and on the physician-patient relationship.

A Longitudinal View of Apathy and Its Impact After Stroke

Background and Purpose— Stroke survivors are often described as apathetic. Because apathy may be a barrier to participation in promising therapies, more needs to be learned about apathy symptoms after stroke. The specific objective was to estimate the extent to which apathy changes with time over the first year after stroke and the impact of apathy on recovery. Methods— The Apathy Assessed cohort was formed from stroke survivors participating in a longitudinal study of health-related quality of life after stroke. A family caregiver completed an apathy questionnaire by telephone at 1, 3, 6, and 12 months after stroke (n=408). Group-based trajectory modeling and ordinal regression were used to identify distinctive groups of individuals with similar trajectories of apathy over the first year after stroke and predictors of apathy trajectory. Results— Both 3- and 5-group trajectory models fit the data. We used the 5-group model because of the potential to further explore the apathy construct. The largest group (50%) had low apathy and 33% had minor apathy that remained stable throughout the first year after stroke. A small proportion (3%) of the study sample had high apathy that remained high. Two other groups of almost equal size (7%) showed worsening and improving apathy. Poor cognitive status, low functional status, and high comorbidity predicted higher apathy. High apathy had a significant negative effect on physical function, participation, health perception, and physical health over the first 12 months after stroke. Conclusion— Some degree of apathy was prevalent and persistent after stroke and was predicted by older age, poor cognitive status, and low functional status after stroke. Even a minor level of apathy had an important and statistically significant impact on stroke outcomes.

Costs and Caregiver Consequences of Early Supported Discharge for Stroke Patients

Background and Purpose— Early supported discharge (ESD) for stroke has been shown to yield outcomes similar to or better than those of conventional care, but there is less information on the impact on costs and on the caregiver. The purpose of this study is to estimate the costs associated with an ESD program compared with those of usual care. Methods— We conducted a randomized controlled trial of stroke patients who required rehabilitation services and who had a caregiver at home. Results— Acute-care costs incurred before randomization when patients were medically ready for discharge averaged

The Ste-justine Adolescent Idiopathic Scoliosis Cohort Study: Part Ii

3251 per person. The costs for the balance of the acute-care stay, from randomization to discharge, were

The association between cigarette smoking and back pain in adults

1383 for the home group and

Adverse Reproductive Outcomes among Women Exposed to Low Levels of Ionizing Radiation from Diagnostic Radiography for Adolescent Idiopathic Scoliosis

2220 for the usual care group. The average cost of providing the 4-week home intervention service was

Nonmedical Bed-Days for Stroke Patients Admitted to Acute-Care Hospitals in Montreal, Canada

943 per person. The total cost generated by persons assigned to the home group averaged

A randomized trial of two home-based exercise programmes to improve functional walking post-stroke

7784 per person, significantly lower than the