Effect of a clinical pathway to reduce hospitalizations in nursing home residents with pneumonia: a randomized controlled trial

CONTEXT

Nursing home residents with pneumonia are frequently hospitalized. Such transfers may be associated with multiple hazards of hospitalization as well as economic costs.

OBJECTIVE

To assess whether using a clinical pathway for on-site treatment of pneumonia and other lower respiratory tract infections in nursing homes could reduce hospital admissions, related complications, and costs.

DESIGN, SETTING, AND PARTICIPANTS

A cluster randomized controlled trial of 680 residents aged 65 years or older in 22 nursing homes in Hamilton, Ontario, Canada. Nursing homes began enrollment between January 2, 2001, and April 18, 2002, with the last resident follow-up occurring July 4, 2005. Residents were eligible if they met a standardized definition of lower respiratory tract infection.

INTERVENTIONS

Treatment in nursing homes according to a clinical pathway, which included use of oral antimicrobials, portable chest radiographs, oxygen saturation monitoring, rehydration, and close monitoring by a research nurse, or usual care.

MAIN OUTCOME MEASURES

Hospital admissions, length of hospital stay, mortality, health-related quality of life, functional status, and cost.

RESULTS

Thirty-four (10%) of 327 residents in the clinical pathway group were hospitalized compared with 76 (22%) of 353 residents in the usual care group. Adjusting for clustering of residents in nursing homes, the weighted mean reduction in hospitalizations was 12% (95% confidence interval [CI], 5%-18%; P = .001). The mean number of hospital days per resident was 0.79 in the clinical pathway group vs 1.74 in the usual care group, with a weighted mean difference of 0.95 days per resident (95% CI, 0.34-1.55 days; P = .004). The mortality rate was 8% (24 deaths) in the clinical pathway group vs 9% (32 deaths) in the usual care group, with a weighted mean difference of 2.9% (95% CI, -2.0% to 7.9%; P = .23). There were no significant differences between the groups in health-related quality of life or functional status. The clinical pathway resulted in an overall cost savings of US 1016 dollars per resident (95% CI, 207 dollars-1824 dollars) treated.

CONCLUSION

Treating residents of nursing homes with pneumonia and other lower respiratory tract infections with a clinical pathway can result in comparable clinical outcomes, while reducing hospitalizations and health care costs.

TRIAL REGISTRATION

clinicaltrials.gov Identifier: NCT00157612.

Stability of the gross motor function classification system

The aim of this study was to assess the stability of the Gross Motor Function Classification System (GMFCS) by examining whether children with cerebral palsy (CP) remain in the same level over time. Participants were 610 children with CP (342 males, 268 females; mean age 6y 9mo [SD 2y 10mo]), range 16mo-13y). Children were assessed 2 to 7 times (mean 4.3) at 6-month (children <6y old) or 12-month(children >or=6y old) intervals. Seventy-three per cent of children remained in the same level for all ratings. The weighted kappa coefficient between the first and last ratings was 0.84 for children less than 6 years old and 0.89 for children at least 6 years old, indicating excellent chance-corrected agreement. Children initially classified in Levels I and V were least likely to be reclassified. There was a tendency for children younger than 6 years who were reclassified to be done so to a lower level of ability. The results provide evidence of stability of the GMFCS.

Interobserver reliability of radiologists' interpretations of mobile chest radiographs for nursing home-acquired pneumonia

OBJECTIVES

To determine the interobserver reliability of radiologists' interpretations of mobile chest radiographs for nursing home-acquired pneumonia.

DESIGN

A cross-sectional reliability study.

SETTING

Nursing homes and an acute care hospital.

PARTICIPANTS

Four radiologists reviewed 40 mobile chest radiographs obtained from residents of nursing homes who met a clinical definition of lower respiratory tract infections.

MEASUREMENTS

Radiologists were asked to interpret radiographs with respect to the film quality; presence, pattern, and extent of an infiltrate; and the presence of a pleural effusion or adenopathy. Interrater reliability was evaluated using the intraclass correlation coefficient derived from a 2-way random effects model.

RESULTS

On average the radiologists reported that 6 of the 40 films were of very good or excellent quality and 16 of the 40 were of fair or poor quality. When the finding of an infiltrate was dichotomized (0 = no; 1 = possible, probable, or definite) all 4 radiologists agreed on 21 of the 37 chest radiographs. The intraclass correlation coefficient for the presence or absence of infiltrates was 0.54 (95% confidence intervals [CI] 0.38 to 0.69). For the 14 radiographs where infiltrates were observed by all radiologists, intraclass correlation coefficients for the presence of pleural effusions was 0.08 (95% CI -0.10 to 0.41), hilar adenopathy 0.54 (95% CI 0.29 to 0.79), and mediastinal adenopathy 0.49 (95% CI 0.21 to 0.76).

CONCLUSION

In conclusion, the interrater agreement among radiologists for mobile chest radiographs in establishing the presence or absence of an infiltrate can be judged to be "fair." Treatment decisions need to include clinical findings and should not be made based on radiographic findings alone.

Interval estimation and optimal design for the within-subject coefficient of variation for continuous and binary variables

BACKGROUND

In this paper we propose the use of the within-subject coefficient of variation as an index of a measurement's reliability. For continuous variables and based on its maximum likelihood estimation we derive a variance-stabilizing transformation and discuss confidence interval construction within the framework of a one-way random effects model. We investigate sample size requirements for the within-subject coefficient of variation for continuous and binary variables.

METHODS

We investigate the validity of the approximate normal confidence interval by Monte Carlo simulations. In designing a reliability study, a crucial issue is the balance between the number of subjects to be recruited and the number of repeated measurements per subject. We discuss efficiency of estimation and cost considerations for the optimal allocation of the sample resources. The approach is illustrated by an example on Magnetic Resonance Imaging (MRI). We also discuss the issue of sample size estimation for dichotomous responses with two examples.

RESULTS

For the continuous variable we found that the variance stabilizing transformation improves the asymptotic coverage probabilities on the within-subject coefficient of variation for the continuous variable. The maximum like estimation and sample size estimation based on pre-specified width of confidence interval are novel contribution to the literature for the binary variable.

CONCLUSION

Using the sample size formulas, we hope to help clinical epidemiologists and practicing statisticians to efficiently design reliability studies using the within-subject coefficient of variation, whether the variable of interest is continuous or binary.

Mortality predictions in the intensive care unit: comparing physicians with scoring systems

OBJECTIVE

Risk-prediction models offer potential advantages over physician predictions of outcomes in the intensive care unit (ICU). Our systematic review compared the accuracy of ICU physicians' and scoring system predictions of ICU or hospital mortality of critically ill adults.

DATA SOURCE

MEDLINE (1966-2005), CINAHL (1982-2005), Ovid Healthstar (1975-2004), EMBASE (1980-2005), SciSearch (1980-2005), PsychLit (1985-2004), the Cochrane Library (Issue 1, 2005), PubMed "related articles," personal files, abstract proceedings, and reference lists.

STUDY SELECTION

We considered all studies that compared physician predictions of ICU or hospital survival of critically ill adults to an objective scoring system, computer model, or prediction rule. We excluded studies if they focused exclusively on the development or economic evaluation of a scoring system, computer model, or prediction rule.

DATA EXTRACTION AND ANALYSIS

We independently abstracted data and assessed study quality in duplicate. We determined summary receiver operating characteristic curves and areas under the summary receiver operating characteristic curves+/-se and summary diagnostic odds ratios.

DATA SYNTHESIS

We included 12 observational studies of moderate methodological quality. The area under the summary receiver operating characteristic curves for seven studies was 0.85+/-0.03 for physician predictions compared with 0.63+/-0.06 for scoring system predictions (p=.002). Physicians' summary diagnostic odds ratios derived from the area under the summary receiver operating characteristic curves were significantly higher (12.43; 95% confidence interval 5.47, 27.11) than scoring systems' summary diagnostic odds ratios (2.25; 95% confidence interval 0.78, 6.52, p=.001). Combined results of all 12 studies indicated that physicians predict mortality more accurately than do scoring systems: ratio of diagnostic odds ratios (95% confidence interval) 1.92 (1.19, 3.08) (p=.007).

CONCLUSIONS

Observational studies suggest that ICU physicians discriminate between survivors and nonsurvivors more accurately than do scoring systems in the first 24 hrs of ICU admission. The overall accuracy of both predictions of patient mortality was moderate, implying limited usefulness of outcome prediction in the first 24 hrs for clinical decision making.

A new preference-based analysis for randomized trials can estimate treatment acceptability and effect in compliant patients

BACKGROUND AND OBJECTIVES

Development of a new method of analysis to evaluate the acceptability of (or preferences for) the treatments in a randomized trial, and the benefit of treatment among compliers.

MATERIALS AND METHODS

We characterize trial participants through the groups who would: accept either treatment if offered (compliers); refuse one treatment but accept the other if it is offered to them (two groups of preferers); or prefer one treatment and insist on it if it is not offered to them initially (two groups of insisters).

RESULTS

We show that in our framework, one can always estimate the proportions of patients in these five preference groups. However, constraints are required to estimate the corresponding outcome rates, and thus estimate the treatment effect in the compliers. We propose two possible sets of constraints and illustrate them by numerical examples.

CONCLUSIONS

The traditional intention-to-treat analysis avoids biases associated with the alternative per-protocol or as-treated approaches, but it provides imperfect information about the expected treatment effect among patients who are committed to taking the treatment. Many physicians and patients want to know the expected benefit if they adhere to the therapy. Our preference-based analysis provides an estimate of treatment benefit among such patients.

New methods give better estimates of changes in diagnostic accuracy when prior information is provided

OBJECTIVE

Whether tests such as imaging should be read with or without access to prior clinical information is controversial. Naïve comparisons may suggest that the provision of prior information improves test accuracy, whereas in fact the opposite may be true. This is because provision of clinical background may actually bias test readers to over- or underinterpret relevant test findings, and they may suboptimally integrate the previous and current evidence. We propose comparing the combined accuracy of prior information and a test read (i) with or (ii) without knowledge of prior information. Analysis methods include simple decision rules and logistic regression.

STUDY DESIGN AND SETTING

A study of cancer detection in women presenting with breast symptoms, in whom ultrasound could be read with or without reviewing prior mammography.

RESULTS

Naïve analysis gave an area under the receiver operating characteristics curve (AUC) for ultrasound read with mammography on view that was 4.6% higher (P < .01) than without mammography on view. Our approach, comparing the combined accuracy of mammography and ultrasound read i) with and ii) without knowledge of mammographic findings, showed much smaller differences.

CONCLUSION

Our approach is more appropriate than naïve analyses. The particular choice of analytic method depends on the study size and the diagnostic accuracy of combinations of the prior information and the test reading.

The partial area under the summary ROC curve

The area under the curve (AUC) is commonly used as a summary measure of the receiver operating characteristic (ROC) curve. It indicates the overall performance of a diagnostic test in terms of its accuracy at various diagnostic thresholds used to discriminate cases and non-cases of disease. The AUC measure is also used in meta-analyses, where each component study provides an estimate of the test sensitivity and specificity. These estimates are then combined to calculate a summary ROC (SROC) curve which describes the relationship between-test sensitivity and specificity across studies. The partial AUC has been proposed as an alternative measure to the full AUC. When using the partial AUC, one considers only those regions of the ROC space where data have been observed, or which correspond to clinically relevant values of test sensitivity or specificity. In this paper, we extend the idea of using the partial AUC to SROC curves in meta-analysis. Theoretical and numerical results describe the variation in the partial AUC and its standard error as a function of the degree of inter-study heterogeneity and of the extent of truncation applied to the ROC space. A scaled partial area measure is also proposed to restore the property that the summary measure should range from 0 to 1. The results suggest several disadvantages of the partial AUC measures. In contrast to earlier findings with the full AUC, the partial AUC is rather sensitive to heterogeneity. Comparisons between tests are more difficult, especially if an empirical truncation process is used. Finally, the partial area lacks a useful symmetry property enjoyed by the full AUC. Although the partial AUC may sometimes have clinical appeal, on balance the use of the full AUC is preferred.

Effect of a multifaceted intervention on number of antimicrobial prescriptions for suspected urinary tract infections in residents of nursing homes: cluster randomised controlled trial

OBJECTIVE

To assess whether a multifaceted intervention can reduce the number of prescriptions for antimicrobials for suspected urinary tract infections in residents of nursing homes.

DESIGN

Cluster randomised controlled trial.

SETTING

24 nursing homes in Ontario, Canada, and Idaho, United States.

PARTICIPANTS

12 nursing homes allocated to a multifaceted intervention and 12 allocated to usual care. Outcomes were measured in 4217 residents.

INTERVENTIONS

Diagnostic and treatment algorithm for urinary tract infections implemented at the nursing home level using a multifaceted approach--small group interactive sessions for nurses, videotapes, written material, outreach visits, and one on one interviews with physicians.

MAIN OUTCOME MEASURES

Number of antimicrobials prescribed for suspected urinary tract infections, total use of antimicrobials, admissions to hospital, and deaths.

RESULTS

Fewer courses of antimicrobials for suspected urinary tract infections per 1000 resident days were prescribed in the intervention nursing homes than in the usual care homes (1.17 v 1.59 courses; weighted mean difference -0.49, 95% confidence intervals -0.93 to -0.06). Antimicrobials for suspected urinary tract infection represented 28.4% of all courses of drugs prescribed in the intervention nursing homes compared with 38.6% prescribed in the usual care homes (weighted mean difference -9.6%, -16.9% to -2.4%). The difference in total antimicrobial use per 1000 resident days between intervention and usual care groups was not significantly different (3.52 v 3.93; weighted mean difference -0.37, -1.17 to 0.44). No significant difference was found in admissions to hospital or mortality between the study arms.

CONCLUSION

A multifaceted intervention using algorithms can reduce the number of antimicrobial prescriptions for suspected urinary tract infections in residents of nursing homes.

Procedures that assess inconsistency in meta-analyses can assess the likelihood of response bias in multiwave surveys

BACKGROUND AND OBJECTIVE

Response bias may affect the result of surveys with <100% response rate. We applied methods commonly used in meta-analysis to ascertain the extent to which response bias affects multiwave survey results.

METHODS

To test hypotheses of between-wave similarity, we used the Cochran-Armitage test for trends and the Q-test of heterogeneity across waves in a survey of 2,127 North American clinicians using six e-mail waves and one fax wave and achieving a response rate of 22%. We used the I2 statistic To quantify the extent of inconsistency in survey outcomes across waves not due to within-wave random error (i.e., inconsistency due to response bias).

RESULTS

With this survey, tests of heterogeneity and trend were not significant and I2 equaled 0%. These results suggest that the underlying responses did not differ across waves and thus strengthened the inference that response bias was not affecting the interpretation of the survey.

CONCLUSION

Researchers can use procedures that assess inconsistency in meta-analyses to evaluate the validity of a multiwave survey with a less than optimal response rate.

The health and well-being of caregivers of children with cerebral palsy

OBJECTIVE

Most children enjoy healthy childhoods with little need for specialized health care services. However, some children experience difficulties in early childhood and require access to and utilization of considerable health care resources over time. Although impaired motor function is the hallmark of the cerebral palsy (CP) syndromes, many children with this development disorder also experience sensory, communicative, and intellectual impairments and may have complex limitations in self-care functions. Although caregiving is a normal part of being the parent of a young child, this role takes on an entirely different significance when a child experiences functional limitations and possible long-term dependence. One of the main challenges for parents is to manage their child's chronic health problems effectively and juggle this role with the requirements of everyday living. Consequently, the task of caring for a child with complex disabilities at home might be somewhat daunting for caregivers. The provision of such care may prove detrimental to both the physical health and the psychological well-being of parents of children with chronic disabilities. It is not fully understood why some caregivers cope well and others do not. The approach of estimating the "independent" or "direct" effects of the care recipient's disability on the caregiver's health is of limited value because (1) single-factor changes are rare outside the context of constrained experimental situations; (2) assumptions of additive relationships and perfect measurements rarely hold; and (3) such approaches do not provide a complete perspective, because they fail to examine indirect pathways that occur between predictor variables and health outcomes. A more detailed analytical approach is needed to understand both direct and indirect effects simultaneously. The primary objective of the current study was to examine, within a single theory-based multidimensional model, the determinants of physical and psychological health of adult caregivers of children with CP.

METHODS

We developed a stress process model and applied structural equation modeling with data from a large cohort of caregivers of children with CP. This design allowed the examination of the direct and indirect relationships between a child's health, behavior and functional status, caregiver characteristics, social supports, and family functioning and the outcomes of caregivers' physical and psychological health. Families (n = 468) of children with CP were recruited from 19 regional children's rehabilitation centers that provide outpatient disability management and supports in Ontario, Canada. The current study drew on a population available to the investigators from a previous study, the Ontario Motor Growth study, which explored patterns of gross motor development in children with CP. Data on demographic variables and caregivers' physical and psychological health were assessed using standardized, self-completed parent questionnaires as well as a face-to-face home interview. Structural equation modeling was used to test specific hypotheses outlined in our conceptual model. This analytic approach involved a 2-step process. In the first step, observed variables that were hypothesized to measure the underlying constructs were tested using confirmatory factor analysis; this step led to the so-called measurement model. The second step tested hypotheses about relationships among the variables in the structural model. All of the hypothesized paths in the conceptual model were tested and included in the structural model. However, only paths that were significant were shown in the final results. The direct, indirect, and total effects of theoretical constructs on physical and psychological health were calculated using the structural model.

RESULTS

The most important predictors of caregivers' well-being were child behavior, caregiving demands, and family function. A higher level of behavior problems was associated with lower levels of both psychological (beta = -.22) and physical health (beta = -.18) of the caregivers, whereas fewer child behavior problems were associated with higher self-perception (beta = -.37) and a greater ability to manage stress (beta = -.18). Less caregiving demands were associated with better physical (beta = .23) and psychological (beta = .12) well-being of caregivers, respectively. Similarly, higher reported family functioning was associated with better psychological health (beta = .33) and physical health (beta = .33). Self-perception and stress management were significant direct predictors of caregivers' psychological health but did not directly influence their physical well-being. Caregivers' higher self-esteem and sense of mastery over the caregiving situation predicted better psychological health (beta = .23). The use of more stress management strategies was also associated with better psychological health of caregivers (beta = .11). Gross income (beta = .08) and social support (beta = .06) had indirect overall effects only on psychological health outcome, whereas self-perception (beta = .22), stress management (beta = .09), gross income (beta = .07), and social support (beta = .06) had indirect total effects only on physical health outcomes.

CONCLUSIONS

The psychological and physical health of caregivers, who in this study were primarily mothers, was strongly influenced by child behavior and caregiving demands. Child behavior problems were an important predictor of caregiver psychological well-being, both directly and indirectly, through their effect on self-perception and family function. Caregiving demands contributed directly to both the psychological and the physical health of the caregivers. The practical day-to-day needs of the child created challenges for parents. The influence of social support provided by extended family, friends, and neighbors on health outcomes was secondary to that of the immediate family working closely together. Family function affected health directly and also mediated the effects of self-perception, social support, and stress management. In families of children with CP, strategies for optimizing caregiver physical and psychological health include supports for behavioral management and daily functional activities as well as stress management and self-efficacy techniques. These data support clinical pathways that require biopsychosocial frameworks that are family centered, not simply technical and short-term rehabilitation interventions that are focused primarily on the child. In terms of prevention, providing parents with cognitive and behavioral strategies to manage their child's behaviors may have the potential to change caregiver health outcomes. This model also needs to be examined with caregivers of children with other disabilities.

Optimal search strategies for retrieving scientifically strong studies of treatment from Medline: analytical survey

OBJECTIVE

To develop and test optimal Medline search strategies for retrieving sound clinical studies on prevention or treatment of health disorders.

DESIGN

Analytical survey.

DATA SOURCES

161 clinical journals indexed in Medline for the year 2000.

MAIN OUTCOME MEASURES

Sensitivity, specificity, precision, and accuracy of 4862 unique terms in 18 404 combinations.

RESULTS

Only 1587 (24.2%) of 6568 articles on treatment met criteria for testing clinical interventions. Combinations of search terms reached peak sensitivities of 99.3% (95% confidence interval 98.7% to 99.8%) at a specificity of 70.4% (69.8% to 70.9%). Compared with best single terms, best multiple terms increased sensitivity for sound studies by 4.1% (absolute increase), but with substantial loss of specificity (absolute difference 23.7%) when sensitivity was maximised. When terms were combined to maximise specificity, 97.4% (97.3% to 97.6%) was achieved, about the same as that achieved by the best single term (97.6%, 97.4% to 97.7%). The strategies newly reported in this paper outperformed other validated search strategies except for two strategies that had slightly higher specificity (98.1% and 97.6% v 97.4%) but lower sensitivity (42.0% and 92.8% v 93.1%).

CONCLUSION

New empirical search strategies have been validated to optimise retrieval from Medline of articles reporting high quality clinical studies on prevention or treatment of health disorders.

Pre-trial evaluation of the potential for unblinding in drug trials: a prototype example

Blinding is an important design feature of randomised trials that may reduce bias in the results, compared to the situation where blinding is not possible or is not maintained. The literature provides some guidance for the evaluation of blinding in ongoing or completed studies, but the question of pre-trial assessment of the potential for unblinding has not been addressed. This paper describes the design and analysis of a prototype experiment for the pre-trial assessment of blinding in a drug trial. This work was motivated by a trial using antibiotic therapy, in which the investigators were concerned about the possibility of subjects being able to differentiate active medication from placebo, and thus become unblinded to their treatment assignment. A small experiment was mounted in which participants had to divide a random mixture of tablets into two groups. Statistical methods were developed to calculate the probability of a given number of similar tablets being classified into the same group by chance, with a modification to allow for some participants having constrained their responses to have equal numbers of tablets in each group. Differentiation of tablets by taste (the initial concern of the investigators) was not statistically different from chance. A smaller set of data on differentiation by appearance (a possibility not originally considered) had borderline statistical significance. After reviewing all these results, the investigators decided to proceed with the study without modifying the tablets, in part because subjects in the study would be unlikely to compare the two types of medication side-by-side. Our results suggest that blinding might sometimes be compromised in unexpected ways. Whenever possible, we suggest that similar and larger such experiments be carried out before the trial to assess whether blinding might be compromised. The methods proposed here could easily be adapted to evaluate the results of such experiments.

Intra-class correlation estimates for assessment of vitamin A intake in children

In many community-based surveys, multi-level sampling is inherent in the design. In the design of these studies, especially to calculate the appropriate sample size, investigators need good estimates of intra-class correlation coefficient (ICC), along with the cluster size, to adjust for variation inflation due to clustering at each level. The present study used data on the assessment of clinical vitamin A deficiency and intake of vitamin A-rich food in children in a district in India. For the survey, 16 households were sampled from 200 villages nested within eight randomly-selected blocks of the district. ICCs and components of variances were estimated from a three-level hierarchical random effects analysis of variance model. Estimates of ICCs and variance components were obtained at village and block levels. Between-cluster variation was evident at each level of clustering. In these estimates, ICCs were inversely related to cluster size, but the design effect could be substantial for large clusters. At the block level, most ICC estimates were below 0.07. At the village level, many ICC estimates ranged from 0.014 to 0.45. These estimates may provide useful information for the design of epidemiological studies in which the sampled (or allocated) units range in size from households to large administrative zones.

Need for expertise based randomised controlled trials

Surgical procedures are less likely to be rigorously evidence based than drug treatments because of difficulties with randomisation. Expertise based trials could be the way forward

Adjustment for baseline measurement error in randomized controlled trials induces bias

When estimating the treatment effect in a randomized controlled trial, it is common to have a continuous outcome which is also observed at baseline. These observations are often prone to measurement error, for example due to within-patient variability. Controversy exists in the literature about whether baseline measurement error should be adjusted for in this context. Computer simulations were used to compare the biases in the estimated treatment effect, with and without adjusting for measurement error, and for different levels of observed baseline imbalance. The impacts of sample size (30 per group and 300 per group) and reliability coefficient (0.6, 0.8 and 1) were also assessed. The results show that in randomized controlled trials, the ordinary least squares (OLS) estimator without adjusting for measurement error is unbiased. On the contrary, adjusting for measurement error leads to bias, especially when sample sizes are small and/or measurement error is large. The treatment effect adjusting for measurement error is on average overestimated when the baseline mean of the control group is larger than that of the treated group. It is underestimated when the control group has a smaller baseline mean.

The number needed to treat (NNT) can be adjusted for bias when the outcome is measured with error

BACKGROUND AND OBJECTIVE

We consider the number needed to treat (NNT) when the event of interest is defined by dichotomizing a continuous response at a threshold level. If the response is measured with error, the resulting NNT is biased. We consider methods to reduce this bias.

METHODS

Bias adjustment was studied using simulations in which we varied the distributions of the underlying response and measurement error, including both normal and nonnormal distributions. We studied a maximum likelihood estimate (MLE) based on normality assumptions, and also considered a simulation-extrapolation estimate (SIMEX) without such assumptions. The treatment effect across all potential thresholds was summarized using an NNT threshold curve.

RESULTS

Crude NNT estimation was substantially biased due to measurement error. The MLE performed well under normality, and it continued to perform well with nonnormal measurement error, but when the underlying response was nonnormal the MLE was unacceptably biased and was outperformed by the SIMEX estimate. The simulation results were also reflected in empirical data from a randomized study of cholesterol-lowering therapy.

CONCLUSION

Ignoring measurement error can lead to substantial bias in NNT, which can have an important practical effect on the interpretation of analyses. Analysis methods that adjust for measurement error bias can be used to assess the sensitivity of NNT estimates to this effect.

Nursing home residents and Enterobacteriaceae resistant to third-generation cephalosporins

Limited data identify the risk factors for infection with Enterobacteriaceae resistant to third-generation cephalosporins among residents of long-term-care facilities. Using a nested case-control study design, nursing home residents with clinical isolates of Enterobacteriaceae resistant to third-generation cephalosporins were compared to residents with isolates of Enterobacteriaceae susceptible to third-generation cephalosporins. Data were collected on antimicrobial drug exposure 10 weeks before detection of the isolates, facility-level demographics, hygiene facilities, and staffing levels. Logistic regression models were built to adjust for confounding variables. Twenty-seven case-residents were identified and compared to 85 controls. Exposure to any cephalosporin (adjusted odds ratio [OR] 4.0, 95% confidence interval [CI] 1.2 to13.6) and log percentage of residents using gastrostomy tubes within the nursing home (adjusted OR 3.9, 95% CI 1.3 to 12.0) were associated with having a clinical isolate resistant to third-generation cephalosporins.

The health of primary caregivers of children with cerebral palsy: how does it compare with that of other Canadian caregivers?

BACKGROUND

Caring for any child involves considerable resources, but the demands for these resources are often increased when caring for a child with a disability. These demands have implications for the psychologic and physical health of the caregiver (CG). Although a number of recent trends in health care stress the importance of studying and promoting the health of CGs of children with disabilities, the literature in this area exhibits 2 major weaknesses, ie, most studies draw conclusions from relatively small, potentially biased, clinic-based samples and the majority of work has focused on the psychologic health of CGs, whereas little research has been undertaken to study their physical well-being. The goal of this study was to compare the physical and psychologic health of CGs of children with cerebral palsy (CP) with that of the general population of CGs.

METHODS

Data on the physical and psychologic health of 468 primary CGs of children with CP, drawn from 18 of 19 publicly funded children's rehabilitation centers in Ontario, Canada, were collected with a self-completed questionnaire and a face-to-face interview. Identical items and scales had been administered previously to nationally representative samples of the Canadian population in 2 large-scale Canadian surveys, ie, the National Population Health Survey (NPHS) and the National Longitudinal Study of Children and Youth (NLSCY). Subsamples of those data, restricted to adult residents of the province of Ontario who were parents, allowed a comparison of our sample of CGs of children with CP with parent samples from both the NLSCY (n = 2414) and the NPHS (n = 5549).

OUTCOME MEASURES

Demographic variables included CG age, gender, education, income, and work-related variables. Psychologic health and support variables included social support, family functioning, frequency of contacts, distress, and emotional and cognitive problems. Physical health variables included the number and variety of chronic conditions, vision, hearing, and mobility problems, and experience of pain.

RESULTS

CGs of children with CP had lower incomes than did the general population of CGs (proportion with income over 60,000 dollars: CG: 40.9%; NLSCY: 51.4%), despite the absence of any important differences in education between the 2 samples. Results showed that CGs of children with CP were less likely to report working for pay (CG: 66%; NLSCY: 81.2%), less likely to be engaged in full-time work (CG: 67.5%; NLSCY: 73.2%), and more likely to list caring for their families as their main activity (CG: 37.2%; NLSCY: 28.4%). Measures of support showed no difference in reported social support (CG: mean score: 14.5; SD: 3.4; NLSCY: mean score: 14.3; SD: 2.7) or family functioning (CG: mean score: 8.6; SD: 5.6; NLSCY: mean score: 9.0; SD: 4.9) between the 2 samples, although the CG sample did report a statistically greater number of support contacts (CG: mean score: 4.5; SD: 0.7; NPHS: mean score: 4.2; SD: 0.9). Measures of psychologic health showed greater reported distress (CG: mean score: 4.7; SD: 4.4; NPHS: mean score: 2.2; SD: 2.7), chronicity of distress (CG: mean score: 5.5; SD: 1.4; NPHS: mean score: 5.2; SD: 1.1), emotional problems (CG: 25.3% indicating problems; NPHS: 13.7%), and cognitive problems (CG: 38.8%; NPHS: 14.3%) among CGs of children with CP. They also reported a greater likelihood of a variety of physical problems, including back problems (CG: 35.5% reporting the condition; SE: 2.2%; NLSCY: 12.2%; SE: 0.7%), migraine headaches (CG: 24.2%; SE: 2.0%; NLSCY: 11.2%; SE: 0.7%), stomach/intestinal ulcers (CG: 8.4%; SE: 1.3%; NLSCY: 1.7%; SE: 0.3%), asthma (CG: 15.8%; SE: 1.7%; NLSCY: 6.3%; SE: 0.5%), arthritis/rheumatism (CG: 17.3%; SE: 1.8%; NLSCY: 7.3%; SE: 0.5%), and experience of pain (CG: 28.8%; SE: 2.1%; NPHS: 11.0%; SE: 0.5), as well as a greater overall number of chronic physical conditions (CG: 24.1% reporting no chronic conditions; NLSCY: 55.2%).

CONCLUSIONS

Although many families cope well despite the added challenges of caring for a child with a disability, our findings suggest that the demands of their children's disabilities can explain differences in the health status of parents and that parents of children with CP are more likely to have a variety of physical and psychologic health problems. Many of these findings are consistent with a stress process model, in which stress from caregiving can directly or indirectly affect a variety of measures of health, although some of the findings (asthma and arthritis) seem to strain this hypothesis. Alternate interpretations of these findings include the possibility that parents who are in regular contact with the health care system may have more opportunities to discuss and receive attention for their own health concerns than do comparison adults or that the greater number of health issues reported by CGs is related to the nature of our study, perhaps leading these parents to focus on their health and well-being in more depth than is usually feasible in a population survey. CGs of children with CP also had lower incomes, despite the absence of any important differences in education. The findings are consistent with the idea that the financial burden of caring for a child with a disability results in part from a reduced availability of these parents to work for pay. IMPLICATIONS FOR SERVICE PROVIDERS: Physicians and other health care professionals should be aware of the important relationship between child disability and CG health. Family-centered policies and services that explicitly consider CG health are likely to benefit the well-being of both CGs and their families. Future work should address the extent to which the family-centeredness of services, as experienced by CGs, is associated with better health outcomes for parents and their families.

Payments for care at private for-profit and private not-for-profit hospitals: a systematic review and meta-analysis

BACKGROUND

It has been shown that patients cared for at private for-profit hospitals have higher risk-adjusted mortality rates than those cared for at private not-for-profit hospitals. Uncertainty remains, however, about the economic implications of these forms of health care delivery. Since some policy-makers might still consider for-profit health care if expenditure savings were sufficiently large, we undertook a systematic review and meta-analysis to compare payments for care at private for-profit and private not-for-profit hospitals.

METHODS

We used 6 search strategies to identify published and unpublished observational studies that directly compared the payments for care at private for-profit and private not-for-profit hospitals. We masked the study results before teams of 2 reviewers independently evaluated the eligibility of all studies. We confirmed data or obtained additional data from all but 1 author. For each study, we calculated the payments for care at private for-profit hospitals relative to private not-for-profit hospitals and pooled the results using a random effects model.

RESULTS

Eight observational studies, involving more than 350 000 patients altogether and a median of 324 hospitals each, fulfilled our eligibility criteria. In 5 of 6 studies showing higher payments for care at private for-profit hospitals, the difference was statistically significant; in 1 of 2 studies showing higher payments for care at private not-for-profit hospitals, the difference was statistically significant. The pooled estimate demonstrated that private for-profit hospitals were associated with higher payments for care (relative payments for care 1.19, 95% confidence interval 1.07-1.33, p = 0.001).

INTERPRETATION

Private for-profit hospitals result in higher payments for care than private not-for-profit hospitals. Evidence strongly supports a policy of not-for-profit health care delivery at the hospital level.

Limb distribution, motor impairment, and functional classification of cerebral palsy

This study explored the relationships between the Gross Motor Function Classification System (GMFCS), limb distribution, and type of motor impairment. Data used were collected in the Ontario Motor Growth study, a longitudinal cohort study with a population-based sample of children with cerebral palsy (CP) in Canada (n=657; age 1 to 13 years at study onset). The majority (87.8%) of children with hemiplegia were classified as level I. Children with a bilateral syndrome were represented in all GMFCS levels, with most in levels III, IV, and V. Classifications by GMFCS and 'limb distribution' or by GMFCS and 'type of motor impairment' were statistically significantly associated (Pearson's chi2 p<0.001), though the correlation for limb distribution (two categories) by GMFCS was low (tau-b=0.43). An analysis of function (GMFCS) by impairment (limb distribution) indicates that the latter clinical characteristic does not add prognostic value over GMFCS. Although classification of CP by impairment level is useful for clinical and epidemiological purposes, the value of these subgroups as an indicator of mobility is limited in comparison with the classification of severity with the GMFCS.