Measuring the impact of pharmacoepidemiologic research using altmetrics: A case study of a CNODES drug-safety article.

PURPOSE: To provide an overview of altmetrics, including their potential benefits and limitations, how they may be obtained, and their role in assessing pharmacoepidemiologic research impact. METHODS: Our review was informed by compiling relevant literature identified through searching multiple health research databases (PubMed, Embase, and CIHNAHL) and grey literature sources (websites, blogs, and reports). We demonstrate how pharmacoepidemiologists, in particular, may use altmetrics to understand scholarly impact and knowledge translation by providing a case study of a drug-safety study conducted by the Canadian Network of Observational Drug Effect Studies. RESULTS: A common approach to measuring research impact is the use of citation-based metrics, such as an article's citation count or a journal's impact factor. "Alternative" metrics, or altmetrics, are increasingly supported as a complementary measure of research uptake in the age of social media. Altmetrics are nontraditional indicators that capture a diverse set of traceable, online research-related artifacts including peer-reviewed publications and other research outputs (software, datasets, blogs, videos, posters, policy documents, presentations, social media posts, wiki entries, etc). CONCLUSION: Compared with traditional citation-based metrics, altmetrics take a more holistic view of research impact, attempting to capture the activity and engagement of both scholarly and nonscholarly communities. Despite the limited theoretical underpinnings, possible commercial influence, potential for gaming and manipulation, and numerous data quality-related issues, altmetrics are promising as a supplement to more traditional citation-based metrics because they can ingest and process a larger set of data points related to the flow and reach of scholarly communication from an expanded pool of stakeholders. Unlike citation-based metrics, altmetrics are not inherently rooted in the research publication process, which includes peer review; it is unclear to what extent they should be used for research evaluation.

Factors associated with early insulin initiation in Type 2 diabetes: a Canadian cross-sectional study.

AIM: To examine the patient characteristics associated with early initiation of insulin after a diagnosis of Type 2 diabetes. METHODS: We analysed cross-sectional data from the 2012 Canadian Community Health Survey conducted by Statistics Canada. Multivariable logistic regression was used to explore the association between patient sociodemographic and health status characteristics and initiating insulin within 1 year of a diagnosis of Type 2 diabetes (early insulin use). RESULTS: Weighted estimates for the Canadian population showed that 32% of patients with Type 2 diabetes initiated insulin within 1 year of their diagnosis. Of the insulin initiators, 52% were female and 68% were aged ≥60 years. Factors strongly associated with early initiation of insulin were age (60-69 years: adjusted odds ratio 1.89, 95% CI 1.84-1.94; ≥ 70 years, odds ratio 2.08, 95% CI 2.01-2.15, both vs 40-49 years); smoking (smoker vs never: odds ratio 2.39, 95% CI 2.32-2.46); geography (Western Canada: odds ratio 2.75, 95% CI 2.69-2.81; Quebec: odds ratio 2.20, 95% CI 2.13-2.27, both vs Ontario); mental health (poor vs excellent: odds ratio 1.98, 95% CI 1.92-2.04); BMI (overweight vs normal/underweight: odds ratio 1.63, 95% CI 1.58-1.67); oral antidiabetic medication use (yes vs no: odds ratio 0.66, 95% CI 0.65-0.68); and alcohol use (regular vs non-drinker: odds ratio 0.66, 95% CI 0.65-0.68). CONCLUSION: One-third of the study population with Type 2 diabetes initiated insulin within their first year of diagnosis. Age, smoking status, geographical location, mental health, BMI, education, oral antidiabetic medication use, employment, physical activity, language, doctor visits and alcohol consumption were associated with timing of insulin initiation.

Incretin-based medications for type 2 diabetes: an overview of reviews.

AIMS: To summarize evidence from and assess the quality of published systematic reviews evaluating the safety, efficacy and effectiveness of incretin-based medications used in the treatment of type 2 diabetes. METHODS: We identified systematic reviews of randomized controlled trials or observational studies published in any language that evaluated the safety and/or effectiveness of glucagon-like peptide-1 (GLP-1) receptor agonists or dipeptidyl-peptidase-4 (DPP-4) inhibitors. Data sources used include the Cochrane Library, PubMed, EMBASE, Web of Science, International Pharmaceutical Abstracts, table of contents of diabetes journals, and hand-searching of reference lists and clinical practice guidelines. The methodological quality of systematic reviews was independently assessed by two reviewers using the Assessment of Multiple Systematic Reviews (AMSTAR) checklist. Our study protocol was registered with PROSPERO (2013:CRD42013005149). The primary outcomes were pooled treatment effect estimates for glycaemic control, macrovascular and microvascular complications, and hypoglycaemic events. RESULTS: We identified 467 unique citations of which 84 systematic reviews met our inclusion criteria. There were 51 reviews that evaluated GLP-1 receptor agonists and 64 reviews that evaluated DPP-4 inhibitors. The median (interquartile range) AMSTAR score was 6 (3) out of 11 for quantitative and 1 (1) for non-quantitative reviews. Among the 66 quantitative systematic reviews, there were a total of 718 pooled treatment effect estimates reported for our primary outcomes and 1012 reported pooled treatment effect estimates for secondary outcomes. CONCLUSIONS: Clinicians and policy makers, when using the results of systematic reviews to inform decision-making with regard to round clinical care or healthcare policies for incretin-based medications, should consider the variability in quality of reviews.

Risk of acute coronary events associated with glyburide compared with gliclazide use in patients with type 2 diabetes: a nested case-control study.

AIM: Sulfonylureas might increase the risk of adverse cardiovascular events; however, emerging evidence suggests there may be important differences amongst these drugs. Some, like glyburide, inhibit KATP channels in the heart and pancreas, while others, like gliclazide, are more likely to selectively inhibit KATP channels in the pancreas. We hypothesized that the risk of acute coronary syndrome (ACS) events would be higher in patients using glyburide compared with gliclazide. METHODS: This nested case-control study used administrative health data from Alberta, Canada. New users of glyburide or gliclazide aged ≥66 years between 1998 and 2010 were included. Cases were individuals with an ACS-related hospitalization or death. Up to four controls were matched based on birth year, sex, cohort-entry year and follow-up time. Multivariable conditional logistic regression was used to estimate adjusted odds ratios (OR), controlling for baseline drug use and co-morbidities. RESULTS: Our cohort included 7441 gliclazide and 13 884 glyburide users; 51.4% men, mean (s.d.) age 75.5 (6.6) years and mean (s.d.) duration of follow-up 5.5 (4.0) years. A total of 4239 patients had an ACS-related hospitalization or death and were matched to 16 723 controls. Compared with gliclazide use, glyburide use was associated with a higher risk (adjusted OR 1.14; 95% CI 1.06-1.23) of ACS-related hospitalization or death over 5.5 years (number needed to harm: 50). CONCLUSION: In this observational study, glyburide use was associated with a 14% higher risk of ACS events compared with gliclazide use. Although the difference is small and probably to have implications at the population level rather than the individual patient or clinician, any causal inferences regarding sulfonylurea use and adverse cardiovascular risk should be tested in a large-scale randomized controlled trial.

Impact of guideline-concordant antibiotics and macrolide/ß-lactam combinations in 3203 patients hospitalized with pneumonia: prospective cohort study.

For patients hospitalized with pneumonia, guidelines provide empirical antibiotic recommendations and some studies suggest that macrolide/ß-lactam combinations are preferable. We hypothesized that guideline-concordant regimens, particularly macrolide/ß-lactams, would reduce mortality and ICU admissions. All patients hospitalized with pneumonia in Edmonton, Alberta, Canada, were managed according to a clinical pathway and enrolled in a population-based registry. Clinical data, Pneumonia Severity Index and treatments were collected. Guideline-concordant regimens were macrolides/ß-lactams or respiratory fluoroquinolone monotherapy. The main outcome was in-hospital mortality. The study included 3203 patients and most had severe pneumonia (63% PSI Class IV-V). Three hundred and twenty-one (10.0%) patients died, 306 (9.6%) were admitted to the ICU and 570 (17.8%) achieved the composite of death or ICU admission. Most (n = 2506) patients received guideline-concordant antibiotics. Receipt of guideline-concordant antibiotics was not associated with a reduction in mortality alone (231 (9.2%) vs. 90 (12.9%); adjusted odds ratio (aOR), 0.82; 95% CI, 0.61-1.09; p 0.16), but was associated with decreased death or ICU admission (14.7% vs. 29.0%; aOR, 0.44; 95% CI, 0.36-0.54; p < 0.0001). Within guideline-concordant subgroups, there was no difference in mortality between macrolide/ß-lactams and respiratory fluoroquinolone monotherapy (22 (8.3%) vs. 209 (9.3%); aOR, 1.09; 95% CI, 0.66-1.81; p 0.73) but macrolide/ß-lactams were associated with increased odds of death or ICU admission (17.4% vs. 14.4%; aOR, 1.58; 95% CI, 1.09-2.27; p 0.01). In conclusion, guideline-concordant antibiotics were not associated with decreased mortality for patients hospitalized with pneumonia, but were associated with a decrease in the composite endpoint of death or ICU admission. Our findings do not support any clinical advantage of macrolide/ß-lactam compared with respiratory fluoroquinolone monotherapy.

Dysglycaemia and 90 day and 1 year risks of death or readmission in patients hospitalised for community-acquired pneumonia.

AIMS/HYPOTHESIS: The aim of this study was to investigate whether dysglycaemia at admission is associated with adverse events at 90 days or 1 year in a population-based cohort of patients hospitalised with community-acquired pneumonia (CAP). METHODS: Clinical and laboratory data were prospectively collected on all 2,366 adults without diabetes admitted with CAP to six hospitals in Edmonton (AB, Canada) and grouped according to admission glucose: 4.0 to <6.1 mmol/l(n=778, reference group), 6.1 to <7.8 mmol/l (n=924); 7.8 to<11.1 mmol/l (n=535); and 11.1 to 20 mmol/l (n=129). Multivariable Cox models were used to examine the relationship between dysglycaemia and mortality or CAP readmission during follow-up. RESULTS: The mean age was 69 (SD 18) years and 48% of participants were female. Compared with those with glucose <6.1 mmol/l (114 [15%] deaths), no differences in 90 day mortality were observed in the dysglycaemia groups: 143 deaths (15%) in the 6.1-7.8 mmol/l group (adjusted HR [aHR] 0.92, 95% CI 0.72-1.18), 111 deaths (21%) in the 7.8-11.1 mmol/l group (aHR 1.05, 0.81-1.37)and 34 deaths (26%) in the 11.1-20 mmol/l group (aHR 1.30, 0.88-1.93). Similarly, compared with those in the <6.1 mmol/l group (198 [25%] deaths), no difference in 1 year mortality was observed: 233 deaths (25%) in the 6.1 to <7.8 mmol/l group (aHR 0.86, 0.71-1.04), 164 deaths (31%) in the 7.8 to <11.1 mmol/l group (aHR 0.92, 0.75-1.14) and 49 deaths (38%) in the 11.1 to 20 mmol/l group (aHR 1.12, 0.81-1.55). Readmissions for CAP were also similar at 1 year: compared with 10% (70/707) in the 6.1 mmol/l group, the frequencies were 8% (66/842), 9% (45/474) and 10% (11/107) in the 6.1 to <7.8 mmol/l, 7.8 to <11.1 mmol/l, and 11.1 to 20 mmol/l groups, respectively (p>0.05 for all comparisons). CONCLUSIONS/INTERPRETATION: Although previously associated with inpatient morbidity and mortality, admission dysglycaemia was not associated with an increased risk of death or CAP readmission at 90 days or 1 year among those who survived hospitalisation for pneumonia.

Insulin use and increased risk of mortality in type 2 diabetes: a cohort study.

AIM: To compare population-based rates of all-cause and cardiovascular (CV) mortality in newly treated patients with type 2 diabetes according to levels of insulin exposure. METHODS: Using the administrative databases of Saskatchewan Health, 12272 new users of oral antidiabetic therapy were identified between 1991 and 1996 and grouped according to cumulative insulin exposure based on total insulin dispensations per year: no exposure (reference group); low exposure (0 to <3); moderate exposure (3 to <12) and high exposure (> or =12). Time-varying multivariable Cox proportional hazards models were used to examine the relationship between insulin exposure and all-cause, CV-related and non-vascular mortality after adjustment for demographics, medications and comorbidities. RESULTS: Average age was 65 (s.d. 13.9) years, 45% were female, and mean follow-up was 5.1 (s.d. 2.2) years. In total, 1443 (12%) subjects started insulin, and 2681 (22%) deaths occurred. The highest mortality rates were in the high exposure group; 95 deaths/1000 person-years compared with 40 deaths/1000 person-years in the no exposure group [unadjusted hazard ratio (HR): 2.32; 95% confidence interval (CI): 1.96-2.73]. After adjustment, we observed a graded risk of mortality associated with increasing exposure to insulin: low exposure [adjusted HR (aHR): 1.75; 95% CI: 1.24-2.47], moderate exposure (aHR: 2.18; 1.82-2.60) and high exposure (aHR: 2.79; 2.36-3.30); p = 0.005 for trend. Analyses restricted to CV-related (p = 0.042 for trend) and non-vascular (p = 0.004 for trend) mortality showed virtually identical results. CONCLUSIONS: We observed a significant and graded association between mortality risk and insulin exposure level in an inception cohort of patients with type 2 diabetes that persisted despite multivariable adjustment.