Vinay Mehta

Performance-Based or Self-Report Measures of Physical Function : Which Should Be Used in Clinical Trials of Hip Fracture Patients?

OBJECTIVES To assess the validity, sensitivity to change, and responsiveness of 3 self-report and 4 performance-based measures of physical function: activity measure for postacute care (AM-PAC) Physical Mobility and Personal Care scales, the Medical Outcomes Study 36-Item Short Form Health Survey Physical Function scale (SF-36 PF), the Physical Functional Performance test (PFP-10), the Short Physical Performance Battery (SPPB), a 4-meter gait speed, and the six-minute walk test (6MWT). DESIGN A prospective observational study of patients after a hip fracture. Assessments were performed at baseline and 12 weeks postenrollment. SETTING Inpatient and outpatient rehabilitation facilities in Norway, the United Kingdom, Sweden, Israel, Germany, the United States, Denmark, and Spain. PARTICIPANTS A sample of study participants (N=108) who had a hip fracture. INTERVENTIONS Not applicable. MAIN OUTCOME MEASURES Assessments of validity (known-groups, concurrent, construct, and predictive), sensitivity to change (effect size, standardized response mean [SRM], SE of measure, minimal detectable change (MDC), and responsiveness (optimal operating cut-points and area under the curve) between baseline and 12-week follow-up. RESULTS All physical function measures achieved comparably acceptable levels of validity. Odds ratios in predicting patient Global Assessment of Improvement at 12 weeks were as follows: AM-PAC Physical Mobility scale, 5.3; AM-PAC Personal Care scale, 3.6; SF-36 PF, 4.3; SPPB, 2.0; PFP-10, 2.5; gait speed, 1.9; and 6MWT, 2.4. Effect sizes and SRM exceeded 1 SD for all 7 measures. Percent of patients who exceeded the MDC(90) at week 12 were as follows: AM-PAC Physical Mobility scale, 90%; AM-PAC Personal Care scale, 74%; SF-36 PF, 66%; SPPB, 36%; PFP-10, 75%; gait speed, 69%; and 6MWT, 75%. When evaluating responsiveness using the area under receiver operating curves for each measure, all measures had acceptable responsiveness, and no pattern emerged of superior responsiveness depending on the type of measure used. CONCLUSIONS Findings reveal that the validity, sensitivity, and responsiveness of self-report measures of physical function are comparable to performance-based measures in a sample of patients followed after fracturing a hip. From a psychometric perspective, either type of functional measure would be suitable for use in clinical trials where improvement in function is an endpoint of interest. The selection of the most appropriate type of functional measure as the primary endpoint for a clinical trial will depend on other factors, such as the measures feasibility or the strength of the association between the hypothesized mechanism of action of the study intervention and a functional outcome measure.

Association between hypoglycemia and inpatient mortality and length of hospital stay in hospitalized, insulin-treated patients

Abstract Objective: To assess the impact of hypoglycemia on clinical outcomes among hospitalized, insulin-treated patients. Methods: In a retrospective study, hospitalizations in 2005–2007 were identified from a US inpatient electronic medical records database. All encounters for insulin-treated patients with valid blood glucose measurement were included, except for those with a length of stay <24 hours or >30 days. In an encounter-based analysis, associations between hypoglycemic (glucose ≤70 mg/dL) or severe hypoglycemic (glucose ≤50 mg/dL) episodes and inpatient mortality, ischemic events, neurologic complications, and length of stay were evaluated. Results: Among 107,312 admissions, hypoglycemia occurred in 21,561 (20%) and severe hypoglycemia in 7539 (7%). Inpatient mortality occurred in 6.5% of hospitalizations with hypoglycemia and 3.8% of those without (p < 0.001). Inpatient mortality occurred in 7.6% of hospitalizations with a severe hypoglycemic event. Ischemic events (8.1 vs. 8.0%) and neurologic complications (3.8 vs. 3.7%) were similar in hospitalizations with and without a hypoglycemic event, respectively. In multivariate logistic regression analyses adjusting for age, gender, and selected comorbidities, hypoglycemia was associated with a significant increase in inpatient mortality risk (adjusted odds ratio (OR) = 1.66 [95% CI: 1.55, 1.78]). Similar results were observed with severe hypoglycemia (adjusted OR = 1.44 [1.38, 1.52]). Length of stay was increased in hospitalizations with hypoglycemia (median [interquartile range]: 8.2 days [4.9, 13.9] vs. 5.2 days [3.1, 8.3]; p < 0.0001). Limitations: Due to the nature of the data source, some data of interest were not available, including insulin dose and dose regimen, outpatient medical histories (including diabetes history), pre-hospitalization medications, and cause of death. Conclusions: Hypoglycemia was common among hospitalized patients receiving insulin and, while a direct causal relationship cannot be assumed, was associated with an increased risk of inpatient mortality and increased length of hospital stay. Hypoglycemia is an undesirable event and efforts to minimize in-hospital hypoglycemic events are warranted across the spectrum of hospitalized patients.

Antibodies to Toxin B Are Protective Against Clostridium difficile Infection Recurrence

BACKGROUND Although newer studies have evaluated risk factors for recurrent Clostridium difficile infection (CDI), the vast majority did not measure important biomarkers such as endogenous anti-toxin A and anti-toxin B antibody levels. METHODS Data from the placebo group of a phase 2 trial testing monoclonal antibodies to C. difficile toxins A and B for preventing CDI recurrence (rCDI) were analyzed to assess risk factors associated with rCDI. Patients with symptomatic CDI taking metronidazole or vancomycin were enrolled. The primary outcome was rCDI within 84 days of treatment start. Univariate and multivariate logistic regression was used to examine associations between potential risk factors and rCDI. At baseline, demographic and clinical characteristics were recorded; endogenous antibody levels were assessed using 2 enzyme-linked immunosorbent assays. RESULTS A predictor of recurrence was age ≥65 years, and an antibody-mediated immune response to toxin B appears to be protective against rCDI. CONCLUSIONS Our findings demonstrate the importance of clinical as well as immunological risk factors in rCDI and provide more robust evidence for the protective effects of antibody to toxin B in the prevention of rCDI. CLINICAL TRIALS REGISTRATION NCT00350298.

Development and Validation of the Influenza Intensity and Impact Questionnaire (FluiiQ

OBJECTIVE Clinical trials of new agents to reduce the severity and impact of influenza require accurate assessment of the effect of influenza infection. Because there are limited high-quality adult influenza Patient Reported Outcomes (PRO) measures, the aim was to develop and validate a simple but comprehensive questionnaire for epidemiological research and clinical trials. METHODS Construct and item generation was guided by the literature, concept mapping, focus groups, and interviews with individuals with laboratory-confirmed influenza and expert physicians. Items were administered to 311 people with influenza-like illness (ILI) across 25 US sites. Analyses included classic psychometrics, structural equation modeling (SEM), and Rasch analyses. RESULTS Concept mapping generated 149 concepts covering the influenza experience and clustered into symptoms and impact on daily activities, emotions, and others. Items were drafted using simplicity and brevity criteria. Eleven symptoms from the literature underwent review by physicians and patients, and two were removed and one added. The symptoms domain factored into systemic and respiratory symptoms, whereas the impact domains were unidimensional. All domains displayed good internal consistency (Cronbach α ≥ 0.8) except the three-item respiratory domain (α = 0.48). A five-factor SEM indicated excellent fit where systemic, respiratory, and daily activities domains differentiated patients with ILI or confirmed influenza. All scales were responsive over time. CONCLUSIONS Patient and clinician consultations resulted in an influenza PRO measure with high validity and good overall evidence of reliability and responsiveness. The Influenza Intensity and Impact Questionnaire (FluiiQ™) will improve the evaluation of existing and future agents designed to prevent or control influenza infection by increasing the breadth and depth of measurement in this field.

Development and Validation of the RxDx- Dementia Risk Index to Predict Dementia in Patients with Type 2 Diabetes and Hypertension

BACKGROUND Elderly patients with type 2 diabetes mellitus and hypertension are at high risk for developing dementia. In addition to comorbid disease conditions (Dx), prescription drugs (Rx) are important risk factors for dementia. OBJECTIVE Develop and validate the RxDx-Dementia risk index by combining diagnosis and prescription information in a single risk index to predict incident dementia, and compare its performance with diagnosis-based Charlson comorbidity score (CCS) and prescription-based chronic disease score (CDS). METHODS Elderly patients diagnosed with type 2 diabetes mellitus and hypertension, and without prior dementia were identified from the Clinical Practice Research Datalink (2003-2012). A Cox proportional hazard model was constructed to model the time to dementia by incorporating age, gender, and 31 RxDx disease conditions as independent variables. Points were assigned to risk factors to obtain summary risk score. Discrimination and calibration of the risk index were evaluated. Different risk indices were compared against RxDx-Dementia risk index using c-statistic, net reclassification improvement (NRI) and integrated discrimination improvement (IDI). RESULTS Of 133,176 patients with type 2 diabetes mellitus and hypertension, 3.42% patients developed dementia.The c-statistics value for RxDx-Dementia risk index was 0.806 (95% CI, 0.799-0.812). Based on the c-statistics, NRI and IDI values, the RxDx-Dementia risk index performed better compared to CCS, CDS, and their combinations. CONCLUSION The RxDx-Dementia risk index can be a useful tool to identify hypertensive and diabetic patients who are at high risk of developing dementia. This has implications for clinical management of patients with multiple comorbid conditions as well as risk adjustment for database studies.

Regression coefficient-based scoring system should be used to assign weights to the risk index

OBJECTIVE Some previously developed risk scores contained a mathematical error in their construction: risk ratios were added to derive weights to construct a summary risk score. This study demonstrates the mathematical error and derived different versions of the Charlson comorbidity score (CCS) using regression coefficient-based and risk ratio-based scoring systems to further demonstrate the effects of incorrect weighting on performance in predicting mortality. STUDY DESIGN AND SETTING This retrospective cohort study included elderly people from the Clinical Practice Research Datalink. Cox proportional hazards regression models were constructed for time to 1-year mortality. Weights were assigned to 17 comorbidities using regression coefficient-based and risk ratio-based scoring systems. Different versions of CCS were compared using Akaike information criteria (AIC), McFaddens adjusted R2, and net reclassification improvement (NRI). RESULTS Regression coefficient-based models (Beta, Beta10/integer, Beta/Schneeweiss, Beta/Sullivan) had lower AIC and higher R2 compared to risk ratio-based models (HR/Charlson, HR/Johnson). Regression coefficient-based CCS reclassified more number of people into the correct strata (NRI range, 9.02-10.04) compared to risk ratio-based CCS (NRI range, 8.14-8.22). CONCLUSION Previously developed risk scores contained an error in their construction adding ratios instead of multiplying them. Furthermore, as demonstrated here, adding ratios fail to even work adequately from a practical standpoint. CCS derived using regression coefficients performed slightly better than in fitting the data compared to risk ratio-based scoring systems. Researchers should use a regression coefficient-based scoring system to develop a risk index, which is theoretically correct.

Association of Hypoglycemia With Subsequent Dementia in Older Patients With Type 2 Diabetes Mellitus

Background Studies have found conflicting evidence regarding the association of hypoglycemia with dementia. We evaluated an association of hypoglycemia with subsequent dementia in patients with type 2 diabetes. Methods This retrospective longitudinal cohort study used the Clinical Practice Research Datalink, an electronic medical records data from the United Kingdom, from 2003 to 2012. We included patients aged >65 years diagnosed with type 2 diabetes, with no prior diagnosis of dementia. Dementia was defined using diagnosis codes from medical records. All patients were followed from the date of initial diabetes diagnosis. To account for competing risk of death, we used Fine and Grays competing risk model to determine the association of hypoglycemia with dementia while adjusting for potential confounders. Hypoglycemia was modeled as a time-dependent covariate. Results Of 53,055 patients, 5.7% (n = 3,018) had at least one hypoglycemia episodes. The overall incidence rate of dementia was 12.7 per 1,000 person-years. In the fully adjusted model that controlled for all confounders, the occurrence of at least one hypoglycemia episode was associated with 27% higher odds of subsequent dementia (hazard ratio = 1.27; 95% confidence interval = 1.06-1.51). The risk increased with the number of hypoglycemia episodes: one episode (hazard ratio = 1.26; 95% confidence interval = 1.03-1.54); two or more episodes (hazard ratio = 1.50; 95% confidence interval = 1.09-2.08). Conclusions Hypoglycemia is associated with a higher risk of dementia and may be responsible in part for the higher risk of dementia in patients with diabetes. Alternatively, hypoglycemia may be a marker for undiagnosed cognitive impairment, and we cannot rule out the possibility of reverse causation between hypoglycemia and dementia.

Prediction of Recurrent Clostridium Difficile Infection Using Comprehensive Electronic Medical Records in an Integrated Healthcare Delivery System.

BACKGROUND Predicting recurrent Clostridium difficile infection (rCDI) remains difficult. METHODS We employed a retrospective cohort design. Granular electronic medical record (EMR) data had been collected from patients hospitalized at 21 Kaiser Permanente Northern California hospitals. The derivation dataset (2007-2013) included data from 9,386 patients who experienced incident CDI (iCDI) and 1,311 who experienced their first CDI recurrences (rCDI). The validation dataset (2014) included data from 1,865 patients who experienced incident CDI and 144 who experienced rCDI. Using multiple techniques, including machine learning, we evaluated more than 150 potential predictors. Our final analyses evaluated 3 models with varying degrees of complexity and 1 previously published model. RESULTS Despite having a large multicenter cohort and access to granular EMR data (eg, vital signs, and laboratory test results), none of the models discriminated well (c statistics, 0.591-0.605), had good calibration, or had good explanatory power. CONCLUSIONS Our ability to predict rCDI remains limited. Given currently available EMR technology, improvements in prediction will require incorporating new variables because currently available data elements lack adequate explanatory power. Infect Control Hosp Epidemiol 2017;38:1196-1203.

Cost-effectiveness of using an extensively hydrolyzed casein formula containing Lactobacillus rhamnosus GG in managing infants with cow’s milk allergy in the US

Abstract Objective: To estimate the cost-effectiveness of using an extensively hydrolyzed casein formula containing the probiotic Lactobacillus rhamnosus GG (eHCF + LGG; Nutramigen LGG) compared with an eHCF alone and an amino acid formula (AAF) in treating cow’s milk allergy (CMA) in the US, from the perspective of third-party insurers and from parents. Methods: A decision model was used to estimate the probability of cow’s milk allergic infants developing tolerance to cow’s milk by 18 months. The model also estimated the cost to insurers and parents (US dollars at 2016 prices) of managing infants over 18 months after starting one of the formulae, as well as the relative cost-effectiveness of each of the formulae. Results: The probability of developing tolerance to cow’s milk was higher among infants who were fed eHCF + LGG compared with those fed an eHCF alone or an AAF. Infants who are initially fed with eHCF + LGG are expected to utilize fewer healthcare resources than those fed with one of the other formulae. Hence, the estimated total healthcare cost incurred by third-party insurers and parents of initially feeding infants with eHCF + LGG was less than that of feeding infants with an eHCF alone or an AAF. Conclusion: Initial management of newly-diagnosed cows milk allergic infants with eHCF + LGG was found to afford a cost-effective strategy to both third-party insurers and parents when compared to an eHCF alone or an AAF.

Performance of an administrative claims algorithm to estimate the incidence of pure red cell aplasia in chronic hepatitis C patients

Abstract Background: We developed and validated an algorithm to evaluate pure red cell aplasia (PRCA) incidence in chronic hepatitis C (CHC) patients in a large observational database. Methods: