The term 'repeated measures' refers to data with multiple observations on the same sampling unit. In most cases, the multiple observations are taken over time, but they could be over space. It is usually plausible to assume that observations on the same unit are correlated. Hence, statistical analysis of repeated measures data must address the issue of covariation between measures on the same unit. Until recently, analysis techniques available in computer software only o ered the user limited and inadequate choices. One choice was to ignore covariance structure and make invalid assumptions. Another was to avoid the covariance structure issue by analysing transformed data or making adjustments to otherwise inadequate analyses. Ignoring covariance structure may result in erroneous inference, and avoiding it may result in ine cient inference. Recently available mixed model methodology permits the covariance structure to be incorporated into the statistical model. The MIXED procedure of the SAS J System provides a rich selection of covariance structures through the RANDOM and REPEATED statements. Modelling the covariance structure is a major hurdle in the use of PROC MIXED. However, once the covariance structure is modelled, inference about ÿxed e ects proceeds essentially as when using PROC GLM. An example from the pharmaceutical industry is used to illustrate how to choose a covariance structure. The example also illustrates the e ects of choice of covariance structure on tests and estimates of ÿxed e ects. In many situations, estimates of linear combinations are invariant with respect to covariance structure, yet standard errors of the estimates may still depend on the covariance structure.

The term ‘repeated measures’ refers to data with multiple observations on the same sampling unit. In most cases, the multiple observations are taken over time, but they could be over space. It is usually plausible to assume that observations on the same unit are correlated. Hence, statistical analysis of repeated measures data must address the issue of covariation between measures on the same unit. Until recently, analysis techniques available in computer software only offered the user limited and inadequate choices. One choice was to ignore covariance structure and make invalid assumptions. Another was to avoid the covariance structure issue by analysing transformed data or making adjustments to otherwise inadequate analyses. Ignoring covariance structure may result in erroneous inference, and avoiding it may result in inefficient inference. Recently available mixed model methodology permits the covariance structure to be incorporated into the statistical model. The MIXED procedure of the SAS® System provides a rich selection of covariance structures through the RANDOM and REPEATED statements. Modelling the covariance structure is a major hurdle in the use of PROC MIXED. However, once the covariance structure is modelled, inference about fixed effects proceeds essentially as when using PROC GLM. An example from the pharmaceutical industry is used to illustrate how to choose a covariance structure. The example also illustrates the effects of choice of covariance structure on tests and estimates of fixed effects. In many situations, estimates of linear combinations are invariant with respect to covariance structure, yet standard errors of the estimates may still depend on the covariance structure. Copyright © 2000 John Wiley & Sons, Ltd.

No abstract

Urinary incontinence (UI) is a commonly underreported and underdiagnosed condition. The purpose of this trial was to implement and evaluate behavioral management for continence (BMC), an intervention to manage symptoms of UI with older rural women in their homes. Participants were randomized into BMC or a control group, and 178 were followed for between 6 and 24 months. The intervention involved self-monitoring, bladder training, and pelvic muscle exercise with biofeedback. The primary outcome variable-severity of urine loss-was evaluated by pad test. Secondary variables were episodes of urine loss, micturition frequency, voiding interval, quality of life, and subjective report of severity. Urine loss severity at baseline evaluation was not significantly different in the two groups. But using the generalized linear mixed model analysis, at the four follow-ups, severity of urine loss, episodes of urine loss, quality of life, and subjective report of severity were significantly different. At 2 years the BMC group UI severity decreased by 61%; the control group severity increased by 184%. Self-monitoring and bladder training accounted for most of the improvement. The results support the use of simple strategies based on bladder diaries before implementing more complex treatments.

BackgroundClostridium difficile is the most common cause of nosocomial infectious diarrhea in the United States. However, recent reports have documented that C. difficile infections (CDIs) are occurring among patients without traditional risk factors. The purpose of this study was to examine the epidemiology of CA-CDI, by estimating the incidence of CA-CDI and HA-CDI, identifying patient-related risk factors for CA-CDI, and describing adverse health outcomes of CA-CDI.MethodsWe conducted a population-based, retrospective, nested, case-control study within the University of Iowa Wellmark Data Repository from January 2004 to December 2007. We identified persons with CDI, determined whether infection was community-associated (CA) or hospital-acquired (HA), and calculated incidence rates. We collected demographic, clinical, and pharmacologic information for CA-CDI cases and controls (i.e., persons without CDI). We used conditional logistic regression to estimate the odds ratios (ORs) for potential risk factors for CA-CDI.ResultsThe incidence rates for CA-CDI and HA-CDI were 11.16 and 12.1 cases per 100,000 person-years, respectively. CA-CDI cases were more likely than controls to receive antimicrobials (adjusted OR, 6.09 [95% CI 4.59-8.08]) and gastric acid suppressants (adjusted OR, 2.30 [95% CI 1.56-3.39]) in the 180 days before diagnosis. Controlling for other covariates, increased risk for CA-CDI was associated with use of beta-lactam/beta-lactamase inhibitors, cephalosporins, clindamycin, fluoroquinolones, macrolides, and penicillins. However, 27% of CA-CDI cases did not receive antimicrobials in the 180 days before their diagnoses, and 17% did not have any traditional risk factors for CDI.ConclusionsOur study documented that the epidemiology of CDI is changing, with CA-CDI occurring in populations not traditionally considered "high-risk" for the disease. Clinicians should consider this diagnosis and obtain appropriate diagnostic testing for outpatients with persistent or severe diarrhea who have even remote antimicrobial exposure.

Background The research goals of the Cancer Care Outcomes Research and Surveillance (CanCORS) Consortium are to determine how characteristics and beliefs of patients, providers, and health-care organizations influence the treatments and outcomes of individuals with newly diagnosed lung and colorectal cancers. Because CanCORS results will inform national policy, it is important to know how they generalize to the United States population with these cancers. Research Design This study assessed the representativeness of the CanCORS cohort of 10,547 patients with lung cancer (LC) or colorectal cancer (CRC) enrolled between 2003 and 2005. We compared characteristics (gender, race, age and disease stage) to the Surveillance, Epidemiology and End Results (SEER) population of 234,464 patients with new onset of these cancers during the CanCORS recruitment period. Results The CanCORS sample is well matched to the SEER Program for both cancers. In CanCORS, 41% LC / 47% CRC were female versus 47% LC / 49% CRC in SEER. African American, Hispanic and Asian cases differed by no more than 5 percentage points between CanCORS and SEER. The SEER population is slightly older, with the percentage of patients over 75 years 33.1% LC / 37.3% CRC in SEER versus 26.9% LC / 29.4% in CanCORS, and also has a slightly higher proportion of early stage patients. We also found that the CanCORS cohort was representative within specific SEER regions that map closely to CanCORS sites. Conclusions This study demonstrates that the CanCORS Consortium was successful in enrolling a demographically representative sample within the CanCORS regions.

No abstract

Older patients who received chemotherapy had fewer pretherapy events than younger patients and were less likely to receive platinum-based regimens. Nevertheless, older patients had more adverse events during chemotherapy, independent of comorbidity. Potential implicit trade-offs between symptom management and treatment toxicity should be made explicit and additionally studied.