Christopher A. Harle

Prevalence of prediabetes in England from 2003 to 2011: population-based, cross-sectional study

Objective Prediabetes is a high-risk state for developing diabetes and associated complications. The purpose of this paper was to report trends in prevalence of prediabetes for individuals aged 16 and older in England without previously diagnosed diabetes. Setting Data collected by the Health Survey for England (HSE) in England in the years 2003, 2006, 2009 and 2011. Participants Individuals aged 16 and older who participated in the HSE and provided a blood sample. Primary outcome variable Individuals were classified as having prediabetes if glycated haemoglobin was between 5.7% and 6.4% and were not previously diagnosed with diabetes. Results The prevalence rate of prediabetes increased from 11.6% to 35.3% from 2003 to 2011. By 2011, 50.6% of the population who were overweight (body mass index (BMI)>25) and ≥40 years of age had prediabetes. In bivariate relationships, individuals with greater socioeconomic deprivation were more likely to have prediabetes in 2003 (p=0.0008) and 2006 (p=0.0246), but the relationship was not significant in 2009 (p=0.213) and 2011 (p=0.3153). In logistic regressions controlling for age, sex, race/ethnicity, BMI and high blood pressure, the second most socioeconomically deprived had a significantly elevated risk of having prediabetes (2011, OR=1.45; 95% CI 1.26 to 1.88). Conclusions There has been a marked increase in the proportion of adults in England with prediabetes. The socioeconomically deprived are at substantial risk. In the absence of concerted and effective efforts to reduce risk, the number of people with diabetes is likely to increase steeply in coming years.

Quality of health-related online search results

Consumers are increasingly searching for health information online and using that information to inform their decisions and behavior. Because the negative consequences of basing decisions on inaccurate or untrustworthy health information may be particularly serious, it is important to understand the quality of online health information. This study empirically investigates the quality of health information that is returned by popular search engines when queried using a large, comprehensive set of health-related search terms. Findings indicate that a majority of such information returned by popular search engines is of a high quality but quality levels vary across different health topic areas. In particular, searches for terms related to preventive health and social health issues tend to produce lower quality results than terms related to diagnosis and treatment of physical disease or injury. While the overall prevalence of high quality information is greater than that of low quality, the observed variance across health-related terms has important implications for consumers, policy makers, health information providers, and search engines.

Decision support for chronic pain care: how do primary care physicians decide when to prescribe opioids? a qualitative study

BackgroundPrimary care physicians struggle to treat chronic noncancer pain while limiting opioid misuse, abuse, and diversion. The objective of this study was to understand how primary care physicians perceive their decisions to prescribe opioids in the context of chronic noncancer pain management. This question is important because interventions, such as decision support tools, must be designed based on a detailed understanding of how clinicians use information to make care decisions.MethodsWe conducted in-depth qualitative interviews with family medicine and general internal medicine physicians until reaching saturation in emergent themes. We used a funneling approach to ask a series of questions about physicians’ general decision making challenges and use of information when considering chronic opioids. We then used an iterative, open-coding approach to identify and characterize themes in the data.ResultsWe interviewed fifteen physicians with diverse clinical experiences, demographics, and practice affiliations. Physicians said that general decision making challenges in providing pain management included weighing risks and benefits of opioid therapies and time and resource constraints. Also, some physicians described their active avoidance of chronic pain treatment due to concerns about opioid risks. In their decision making, physicians described the importance of objective and consistent information, the importance of identifying “red flags” related to risks of opioids, the importance of information about physical function as an outcome, and the importance of information that engenders trust in patients.ConclusionsThis study identified and described primary care physicians’ struggles to deliver high quality care as they seek and make decisions based on an array of incomplete, conflicting, and often untrusted patient information. Decision support systems, education, and other interventions that address these challenges may alleviate primary care physicians’ struggles and improve outcomes for patients with chronic pain and other challenging conditions.

Effectiveness of Personalized and Interactive Health Risk Calculators A Randomized Trial

Background Risk calculators are popular websites that provide individualized disease risk assessments to the public. Little is known about their effect on risk perceptions and health behavior. Objective This study sought to test whether risk calculator features—namely, personalized estimates of one’s disease risk and feedback about the effects of risk-mitigating behaviors—improve risk perceptions and motivate healthy behavior. Design A web-based experimental study using simple randomization was conducted to compare the effects of 3 prediabetes risk communication websites. Setting The study was conducted in the context of ongoing health promotion activities sponsored by a university’s human resources office. Patients Participants were adult university employees. Intervention The control website presented nonindividualized risk information. The personalized noninteractive website presented individualized risk calculations. The personalized interactive website presented individualized risk calculations and feedback about the effects of hypothetical risk-mitigating behaviors. Measurements Pre- and postintervention risk perceptions were measured in absolute and relative terms. Health behavior was measured by assessing participant interest in follow-up preventive health services. Results On average, risk perceptions decreased by 2%. There was no general effect of personalization or interactivity in aligning subjective risk perceptions with objective risk calculations or in increasing healthy behaviors. However, participants who previously overestimated their risk reduced their perceptions by 16%. This was a significantly larger change than the 2% increase by participants who underestimated their risk. Limitations Results may not generalize to different populations, different diseases, or longer-term outcomes. Conclusions Compared to nonpersonalized information, individualized risk calculators had little positive effect on prediabetes risk perception accuracy or health behavior. Risk perception accuracy was improved in people who receive relatively “good news” about risk rather than “bad news.”

Electronic Medical Record Availability and Primary Care Depression Treatment

ABSTRACTBackgroundElectronic medical records (EMR) are commonly believed to improve quality of care. Primary care patients with multiple chronic conditions have potentially greater opportunity to benefit from receiving care at practices with EMRs if these systems help coordinate complex care.ObjectiveTo examine how chronic conditions impact the odds that depressed patients receive depression treatment in primary care practices with EMRs compared to practices without EMRs.DesignThe study uses logistic regression to analyze cross-sectional data of primary care physician office visits in freestanding, office-based practices from the 2006–2008 National Ambulatory Medical Care Surveys.PatientsAll visits to primary care providers made by patients ages 18 and older with physician-identified depression (N = 3,467).Main MeasuresOutcomes include depression treatment which is defined as receipt or ordering of antidepressant medication and/or mental health counseling.Key ResultsEMRs were associated with significantly lowered odds that depressed patients received depression treatment (OR = 0.75, p = 0.009, 95% CI: 0.61-0.93); however when stratified by the number of chronic conditions, this association was observed only in patients with three or more chronic conditions (OR = 0.50, p > 0.001, 95% CI: 0.36-0.70). EMRs did not have a significant association with depression treatment for patients with two or fewer chronic conditions.ConclusionsEMRs appear to have an unintended negative association with depression care provided during visits made by primary care patients with multiple chronic conditions.

Hospital Characteristics Associated With Achievement of Meaningful Use

EXECUTIVE SUMMARY The objective of this study was to identify factors associated with hospitals that achieved the Medicare meaningful use incentive thresholds for payment under the Health Information Technology for Economic and Clinical Health (HITECH) Act of 2009. We employed a cross‐sectional design using data from the 2011 American Hospital Association Annual Survey, including the Information Technology Supplement; the Centers for Medicare & Medicaid Services report of hospitals receiving meaningful use payments; and the Health Resources and Services Administrations Area Resource File. We used a lagged value from 2010 to determine electronic health record (EHR) adoption. Our methods were a descriptive analysis and logistic regression to examine how various hospital characteristics are associated with the achievement of Medicare meaningful use incentives. Overall, 1,769 (38%) of 4,683 potentially eligible hospitals achieved meaningful use incentive thresholds by the end of 2012. Characteristics associated with organizations that received incentive payments were having an EHR in place in 2010, having a larger bed size, having a single health information technology vendor, obtaining Joint Commission accreditation, operating under for‐profit status, having Medicare share of inpatient days in the middle two quartiles, being eligible for Medicaid incentives, and being located in the Middle Atlantic or South Atlantic census region. Characteristics associated with not receiving incentive payments were being a member of a hospital system and being located in the Mountain or Pacific census region. Thus far, little evidence suggests that the HITECH incentive program has enticed hospitals without an EHR system to adopt meaningful use criteria. Policy makers should consider modifying the incentive program to accelerate the adoption of and meaningful use in hospitals without EHRs.

Measuring Patient Satisfaction's Relationship to Hospital Cost Efficiency: Can Administrators Make a Difference?

Objective: The aim of this study was to assess the ability and means by which hospital administrators can influence patient satisfaction and its impact on costs. Data Sources: Data are drawn from the American Hospital Association’s Annual Survey of Hospitals, federally collected Hospital Cost Reports, and Medicare’s Hospital Compare. Study Design: Stochastic frontier analyses (SFA) are used to test the hypothesis that the patient satisfaction–hospital cost relationship is primarily a latent “management effect.” The null hypothesis is that patient satisfaction measures are main effects under the control of care providers rather than administrators. Principle Findings: Both SFA models were superior to the standard regression analysis when measuring patient satisfaction’s relationship to hospitals’ cost efficiency. The SFA model with patient satisfaction measures treated as main effects, rather than “latent, management effects,” was significantly better comparing the log-likelihood statistics. Higher patient satisfaction scores on the environmental quality and provider communication dimensions were related to lower facility costs. Higher facility costs were positively associated with patients’ overall impressions (willingness to recommend and overall satisfaction), assessments of medication and discharge instructions, and ratings of caregiver responsiveness (pain control and help when called). Conclusions: In the short term, managers have a limited ability to influence patient satisfaction scores, and it appears that working through frontline providers (doctors and nurses) is critical to success. In addition, results indicate that not all patient satisfaction gains are cost neutral and there may be added costs to some forms of quality. Therefore, quality is not costless as is often argued.

School-Located Influenza Vaccination Reduces Community Risk for Influenza and Influenza-Like Illness Emergency Care Visits

Background School-located influenza vaccination (SLIV) programs can substantially enhance the sub-optimal coverage achieved under existing delivery strategies. Randomized SLIV trials have shown these programs reduce laboratory-confirmed influenza among both vaccinated and unvaccinated children. This work explores the effectiveness of a SLIV program in reducing the community risk of influenza and influenza-like illness (ILI) associated emergency care visits. Methods For the 2011/12 and 2012/13 influenza seasons, we estimated age-group specific attack rates (AR) for ILI from routine surveillance and census data. Age-group specific SLIV program effectiveness was estimated as one minus the AR ratio for Alachua County versus two comparison regions: the 12 county region surrounding Alachua County, and all non-Alachua counties in Florida. Results Vaccination of ∼50% of 5–17 year-olds in Alachua reduced their risk of ILI-associated visits, compared to the rest of Florida, by 79% (95% confidence interval: 70, 85) in 2011/12 and 71% (63, 77) in 2012/13. The greatest indirect effectiveness was observed among 0–4 year-olds, reducing AR by 89% (84, 93) in 2011/12 and 84% (79, 88) in 2012/13. Among all non-school age residents, the estimated indirect effectiveness was 60% (54, 65) and 36% (31, 41) for 2011/12 and 2012/13. The overall effectiveness among all age-groups was 65% (61, 70) and 46% (42, 50) for 2011/12 and 2012/13. Conclusion Wider implementation of SLIV programs can significantly reduce the influenza-associated public health burden in communities.

A Clustering Approach to Segmenting Users of Internet-based Risk Calculators

BACKGROUND Risk calculators are widely available Internet applications that deliver quantitative health risk estimates to consumers. Although these tools are known to have varying effects on risk perceptions, little is known about who will be more likely to accept objective risk estimates. OBJECTIVE To identify clusters of online health consumers that help explain variation in individual improvement in risk perceptions from web-based quantitative disease risk information. METHODS A secondary analysis was performed on data collected in a field experiment that measured peoples pre-diabetes risk perceptions before and after visiting a realistic health promotion website that provided quantitative risk information. K-means clustering was performed on numerous candidate variable sets, and the different segmentations were evaluated based on between-cluster variation in risk perception improvement. RESULTS Variation in responses to risk information was best explained by clustering on pre-intervention absolute pre-diabetes risk perceptions and an objective estimate of personal risk. Members of a high-risk overestimater cluster showed large improvements in their risk perceptions, but clusters of both moderate-risk and high-risk underestimaters were much more muted in improving their optimistically biased perceptions. CONCLUSIONS Cluster analysis provided a unique approach for segmenting health consumers and predicting their acceptance of quantitative disease risk information. These clusters suggest that health consumers were very responsive to good news, but tended not to incorporate bad news into their self-perceptions much. These findings help to quantify variation among online health consumers and may inform the targeted marketing of and improvements to risk communication tools on the Internet.

Teaching a Machine to Feel Postoperative Pain: Combining High-Dimensional Clinical Data with Machine Learning Algorithms to Forecast Acute Postoperative Pain.

BACKGROUND Given their ability to process highly dimensional datasets with hundreds of variables, machine learning algorithms may offer one solution to the vexing challenge of predicting postoperative pain. METHODS Here, we report on the application of machine learning algorithms to predict postoperative pain outcomes in a retrospective cohort of 8,071 surgical patients using 796 clinical variables. Five algorithms were compared in terms of their ability to forecast moderate to severe postoperative pain: Least Absolute Shrinkage and Selection Operator (LASSO), gradient-boosted decision tree, support vector machine, neural network, and k-nearest neighbor (k-NN), with logistic regression included for baseline comparison. RESULTS In forecasting moderate to severe postoperative pain for postoperative day (POD) 1, the LASSO algorithm, using all 796 variables, had the highest accuracy with an area under the receiver-operating curve (ROC) of 0.704. Next, the gradient-boosted decision tree had an ROC of 0.665 and the k-NN algorithm had an ROC of 0.643. For POD 3, the LASSO algorithm, using all variables, again had the highest accuracy, with an ROC of 0.727. Logistic regression had a lower ROC of 0.5 for predicting pain outcomes on POD 1 and 3. CONCLUSIONS Machine learning algorithms, when combined with complex and heterogeneous data from electronic medical record systems, can forecast acute postoperative pain outcomes with accuracies similar to methods that rely only on variables specifically collected for pain outcome prediction.