Susan E. Spratt

A comparison of phenotype definitions for diabetes mellitus

OBJECTIVE This study compares the yield and characteristics of diabetes cohorts identified using heterogeneous phenotype definitions. MATERIALS AND METHODS Inclusion criteria from seven diabetes phenotype definitions were translated into query algorithms and applied to a population (n=173 503) of adult patients from Duke University Health System. The numbers of patients meeting criteria for each definition and component (diagnosis, diabetes-associated medications, and laboratory results) were compared. RESULTS Three phenotype definitions based heavily on ICD-9-CM codes identified 9-11% of the patient population. A broad definition for the Durham Diabetes Coalition included additional criteria and identified 13%. The electronic medical records and genomics, NYC A1c Registry, and diabetes-associated medications definitions, which have restricted or no ICD-9-CM criteria, identified the smallest proportions of patients (7%). The demographic characteristics for all seven phenotype definitions were similar (56-57% women, mean age range 56-57 years).The NYC A1c Registry definition had higher average patient encounters (54) than the other definitions (range 44-48) and the reference population (20) over the 5-year observation period. The concordance between populations returned by different phenotype definitions ranged from 50 to 86%. Overall, more patients met ICD-9-CM and laboratory criteria than medication criteria, but the number of patients that met abnormal laboratory criteria exclusively was greater than the numbers meeting diagnostic or medication data exclusively. DISCUSSION Differences across phenotype definitions can potentially affect their application in healthcare organizations and the subsequent interpretation of data. CONCLUSIONS Further research focused on defining the clinical characteristics of standard diabetes cohorts is important to identify appropriate phenotype definitions for health, policy, and research.

Optimizing hospital use of intravenous insulin therapy: improved management of hyperglycemia and error reduction with a new nomogram.

OBJECTIVE To assess the efficacy and safety of intravenous (IV) insulin administration with use of our institutions old protocol (pre-nomogram phase) as compared with our new insulin nomogram (post-nomogram phase), which titrates insulin dose based on the rate of change of plasma glucose values and uses multipliers to determine the new insulin infusion rate. METHODS Hospitalized adults receiving an IV insulin infusion in our tertiary care medical center were enrolled in this study after informed consent was obtained. The study was an observational analysis conducted before and after implementation of the new insulin infusion nomogram. Measurements included episodes of hypoglycemia and incidence of the following errors in the insulin infusion process: (1) episodes of documented failure to increase insulin infusion rate despite persistent hyperglycemia and (2) number of times the IV infusion was stopped without subcutaneous administration of insulin. RESULTS Overall, 66 patients were analyzed (38 in the pre-nomogram phase and 28 in the post-nomogram phase). The new nomogram reduced by nearly 3-fold (from 0.89 +/- 0.68 to 0.36 +/- 0.49 occurrence per patient per 24 hours; P<0.001) the mean incidence of failure to give insulin subcutaneously before discontinuation of IV insulin infusion. Moreover, the nomogram nearly eliminated the error of caregiver nonresponsiveness to persistent hyperglycemia: mean incidence 0.39 +/- 0.65 occurrence per patient per 24 hours before implementation of the new nomogram versus 0.02 +/- 0.09 afterward (P<0.002). There was no statistically significant difference in episodes of hypoglycemia between the 2 study groups. CONCLUSION Safe IV administration of insulin through error prevention is essential. Implementation of a new IV insulin infusion nomogram, which adjusts insulin infusion using multipliers, reduces errors and improves glycemic control without increasing hypoglycemic episodes.

Osteoporosis in lung transplant candidates compared to matched healthy controls

Lakey WC, Spratt S, Vinson EN, Gesty‐Palmer D, Weber T, Palmer S. Osteoporosis in lung transplant candidates compared to matched healthy controls.
Clin Transplant 2011: 25: 426–435.

Methods and initial findings from the Durham Diabetes Coalition: Integrating geospatial health technology and community interventions to reduce death and disability

Objective The Durham Diabetes Coalition (DDC) was established in response to escalating rates of disability and death related to type 2 diabetes mellitus, particularly among racial/ethnic minorities and persons of low socioeconomic status in Durham County, North Carolina. We describe a community-based demonstration project, informed by a geographic health information system (GHIS), that aims to improve health and healthcare delivery for Durham County residents with diabetes. Materials and Methods A prospective, population-based study is assessing a community intervention that leverages a GHIS to inform community-based diabetes care programs. The GHIS integrates clinical, social, and environmental data to identify, stratify by risk, and assist selection of interventions at the individual, neighborhood, and population levels. Results The DDC is using a multifaceted approach facilitated by GHIS to identify the specific risk profiles of patients and neighborhoods across Durham County. A total of 22,982 patients with diabetes in Durham County were identified using a computable phenotype. These patients tended to be older, female, African American, and not covered by private health insurance, compared with the 166,041 persons without diabetes. Predictive models inform decision-making to facilitate care and track outcomes. Interventions include: 1) neighborhood interventions to improve the context of care; 2) intensive team-based care for persons in the top decile of risk for death or hospitalization within the coming year; 3) low-intensity telephone coaching to improve adherence to evidence-based treatments; 4) county-wide communication strategies; and 5) systematic quality improvement in clinical care. Conclusions To improve health outcomes and reduce costs associated with type 2 diabetes, the DDC is matching resources with the specific needs of individuals and communities based on their risk characteristics.

Assessing electronic health record phenotypes against gold-standard diagnostic criteria for diabetes mellitus.

Objective: We assessed the sensitivity and specificity of 8 electronic health record (EHR)-based phenotypes for diabetes mellitus against gold-standard American Diabetes Association (ADA) diagnostic criteria via chart review by clinical experts. Materials and Methods: We identified EHR-based diabetes phenotype definitions that were developed for various purposes by a variety of users, including academic medical centers, Medicare, the New York City Health Department, and pharmacy benefit managers. We applied these definitions to a sample of 173 503 patients with records in the Duke Health System Enterprise Data Warehouse and at least 1 visit over a 5-year period (2007–2011). Of these patients, 22 679 (13%) met the criteria of 1 or more of the selected diabetes phenotype definitions. A statistically balanced sample of these patients was selected for chart review by clinical experts to determine the presence or absence of type 2 diabetes in the sample. Results: The sensitivity (62–94%) and specificity (95–99%) of EHR-based type 2 diabetes phenotypes (compared with the gold standard ADA criteria via chart review) varied depending on the component criteria and timing of observations and measurements. Discussion and Conclusions: Researchers using EHR-based phenotype definitions should clearly specify the characteristics that comprise the definition, variations of ADA criteria, and how different phenotype definitions and components impact the patient populations retrieved and the intended application. Careful attention to phenotype definitions is critical if the promise of leveraging EHR data to improve individual and population health is to be fulfilled.

Effects of pregabalin on heart rate variability in patients with painful diabetic neuropathy.

Many studies have demonstrated that low heart rate variability (HRV) is a risk for high mortality and morbidity in patients with cardiovascular diseases. The primary purpose of the study was to evaluate whether pregabalin improves HRV in patients with diabetes and painful peripheral neuropathy. Resting heart rates were collected by using the LifeShirt System, developed by VivoMetrics (Ventura, Calif), at baseline and at the end of a 4-week intervention of pregabalin or placebo in patients with painful diabetic peripheral neuropathy. Heart rate variability analysis was performed on the collected R-R intervals using the Vivo- VMLA-036-00 3 Logic of the LifeShirt system. Of the 40 patients enrolled in the study, 70% completed the end of 4-week assessments (n = 15 in pregabalin and n = 14 in placebo). Compared with placebo, pregabalin treatment resulted in significant improvement in HRV measured by frequency domain analysis, that is, a reduction in low frequency-high frequency ratio (−1.30 ± 2.89 vs 0.37 ± 0.33, P = 0.03) and power of normalized low frequency (−0.049 ± 0.092 vs 0.0066 ± 0.023, P = 0.02), as well as an increase in power of normalized high frequency (0.039 ± 0.094 vs −0.038 ± 0.066, P = 0.02). Furthermore, pregabalin resulted in greater reduction of pain and symptoms of anxiety and greater improvement of quality of life. The improvement of HRV measures were not correlated with change of those measures. In conclusion, 4-week pregabalin treatment improved HRV in patients with painful diabetic peripheral neuropathy. Trial Registration: NCT00573261 (clinicaltrials.gov)Abbreviations: HRV - heart rate variability, VAS - visual analog scale, BID - twice daily, RSA - respiratory sinus arrhythmia, VLF - very low frequency, LF - low frequency, HF - high frequency, ANN - mean of all R-R intervals, SDNN - standard deviation of R-R intervals, RMSSD - root mean square of successive differences, SDANN - standard deviation of the averages of R-R intervals for all 5-minute segments within the block, PNN50 - number of N-N intervals that differed by more than 50 milliseconds from adjacent intervals divided by the total number of all N-N intervals, NPS - Neuropathic Pain Scale, BPI - Modified Brief Pain Inventory-Short Form, STAI - State-Trait Anxiety Inventory Scale, BDI - Beck Depression Inventory

The implementation and evaluation of an evidence-based protocol to treat diabetic ketoacidosis: a quality improvement study.

This retrospective observational quality improvement study was conducted to determine whether an evidence-based protocol for the treatment of diabetic ketoacidosis improved patient outcomes in our academic medical center. This study evaluated fidelity of providers to the protocol, as well as time to resolution of diabetic ketoacidosis as measured by closure of the anion gap (AG). Other secondary outcomes included time to intravenous fluids, time to potassium replacement, and rates of hypoglycemia and hypokalemia.Two cohorts including historical (N = 41) and current (N = 37) were compared to evaluate the effectiveness of the protocol. There were no differences between group demographics at baseline. After implementation of the protocol, 43.2% of patients were treated using full protocol fidelity, 21.6% were treated with partial fidelity, and 35.1% were not treated using the protocol. Although none of the outcomes reached statistical significance, patients in the current group who were treated with full protocol fidelity had an average time to AG closure that was 3 hr less than those who were not treated according to the protocol, and an average time to potassium replacement that was 2 hr less. When comparing the historical cohort with the patients treated with full protocol fidelity, there was improvement in protocol-treated patients in time to AG closure (2 hr), time to dextrose replacement (1.7 hr), and time to potassium replacement (2 hr). The rates of hypokalemia were improved with protocol treatment; 37.5% of protocol-treated patients had hypokalemia as opposed to 63.4% of those not treated according to protocol.Overall, despite the low fidelity in our institution, the protocol promoted evidence-based practice and patients treated according to the protocol had decreased time to treatment outcomes including quicker AG closure, improved intravenous fluids resuscitation, and more accurate and timely electrolyte correction.

I-131, I-123, and F-18 FDG-PET imaging in a patient with diffuse sclerosing variant of papillary thyroid cancer.

Purpose: To compare the sensitivity of I-123 total body iodine (TBI) scan, I-131 TBI, and PET scanning with 2-deoxy-2[F-18]fluoro-d-glucose (FDG-PET) scans for detection of residual/recurrent disease in patients with diffuse sclerosing variant of papillary thyroid cancer. Materials: A 45-year-old woman with status post-thyroidectomy and modified neck dissection showed papillary thyroid carcinoma with a diffuse sclerosing variant and positive lymph nodes. Six weeks after surgery, I-131 TBI and FDG-PET scans showed no residual or metastatic disease. However, clinical suspicion for disease remained, and an I-123 TBI scan was performed. Results: I-123 TBI showed a tiny residual focus in the left thyroid bed. The patient was treated with 150 mCi oral I-131 sodium iodide; posttreatment scan confirmed the presence of residual disease. Follow-up I-123 TBI scans up to 2 1/2 years posttreatment were negative. Conclusion: Total body imaging with I-123 was more sensitive than I-131 TBI scanning for detecting residual or recurrent disease in patients with well-differentiated thyroid cancer presenting with low preoperative thyroglobulin levels.

The agreement of patient-reported versus observed medication adherence in type 2 diabetes mellitus (T2DM).

Objective Medication adherence in type 2 diabetes mellitus (T2DM) improves glycemic control and is associated with reduced adverse clinical events, and accurately assessing adherence assessment is important. We aimed to determine agreement between two commonly used adherence measures—the self-reported Morisky Medication Adherence Scale (MMAS) and direct observation of medication use by nurse practitioners (NPs) during home visits—and determine the relationship between each measure and glycated hemoglobin (HbA1c). Research design and methods We evaluated agreement between adherence measures in the Southeastern Diabetes Initiative (SEDI) prospective clinical intervention home visit cohort, which included high-risk patients (n=430) in 4 SEDI-participating counties. The mean age was 58.7 (SD 11.6) years. The majority were white (n=210, 48.8%), female (n=236, 54.9%), living with a partner (n=316, 74.5%), and insured by Medicare/Medicaid (n=361, 84.0%). Medication adherence was dichotomized to ‘adherent’ or ‘not adherent’ using established cut-points. Inter-rater agreement was evaluated using Cohens κ coefficient. Relationships among adherence measures and HbA1c were evaluated using the Wilcoxon rank-sum test and c-statistics. Results Fewer patients (n=261, 61%) were considered adherent by self-reported MMAS score versus the NP-observed score (n=338; 79%). Inter-rater agreement between the two adherence measures was fair (κ=0.24; 95% CI 0.15 to 0.33; p<0.0001). Higher adherence was significantly associated with lower HbA1c levels for both measures, yet discrimination was weak (c-statistic=0.6). Conclusions Agreement between self-reported versus directly observed medication adherence was lower than expected. Though scores for both adherence measures were significantly associated with HbA1c, neither discriminated well for discrete levels of HbA1c.

Missing signposts on the roadmap to quality: a call to improve medication adherence indicators in data collection for population research

Purpose: Poor adherence to prescribed medicines is associated with increased rates of poor outcomes, including hospitalization, serious adverse events, and death, and is also associated with increased healthcare costs. However, current approaches to evaluation of medication adherence using real-world electronic health records (EHRs) or claims data may miss critical opportunities for data capture and fall short in modeling and representing the full complexity of the healthcare environment. We sought to explore a framework for understanding and improving data capture for medication adherence in a population-based intervention in four U.S. counties. Approach: We posited that application of a data model and a process matrix when designing data collection for medication adherence would improve identification of variables and data accessibility, and could support future research on medication-taking behaviors. We then constructed a use case in which data related to medication adherence would be leveraged to support improved healthcare quality, clinical outcomes, and efficiency of healthcare delivery in a population-based intervention for persons with diabetes. Because EHRs in use at participating sites were deemed incapable of supplying the needed data, we applied a taxonomic approach to identify and define variables of interest. We then applied a process matrix methodology, in which we identified key research goals and chose optimal data domains and their respective data elements, to instantiate the resulting data model. Conclusions: Combining a taxonomic approach with a process matrix methodology may afford significant benefits when designing data collection for clinical and population-based research in the arena of medication adherence. Such an approach can effectively depict complex real-world concepts and domains by “mapping” the relationships between disparate contributors to medication adherence and describing their relative contributions to the shared goals of improved healthcare quality, outcomes, and cost.