Diabetes: how are we diagnosing and initially managing it?

Can internet search engine queries be used to diagnose diabetes? Analysis of archival search data

AIMS

Diabetes is often diagnosed late. This study aimed to assess the possibility for earlier detection of diabetes from search data, using predictive models trained on large-scale data.

METHODS

We extracted all English-language queries made by people in the USA to Bing during 1 year and identified queries containing symptoms of diabetes. We compared the ability of four different prediction models (linear regression, logistic regression, decision tree and random forest) to distinguish between users who stated that they were diagnosed with diabetes and users who did not refer to diabetes or diabetes drugs but queried about at least one of the symptoms.

RESULTS

We identified 11,050 "new diabetes users" who stated they had been diagnosed with diabetes and approximately 11.5 million "control users" who queried about symptoms without querying for terms related to diabetes. Both the logistic regression and the random forest models were able to distinguish between the populations with an area under curve of 0.92 which translates to a positive predictive value of 56% at a false-positive rate of 1%. The model could identify patients up to 240 days before they mentioned being diagnosed.

CONCLUSIONS

Some undiagnosed diabetes patients can be detected accurately according to their symptom queries to a search engine. Such earlier diagnosis, especially in cases of type 1 diabetes, could be clinically meaningful. The ability of search engines to serve as a population-wide screening tool could potentially be improved using additional data provided by users.

Identifying gaps in the continuum of care for cardiovascular disease and diabetes in two communities in South Africa: Baseline findings from the HealthRise project

BACKGROUND

The HealthRise initiative seeks to implement and evaluate innovative community-based strategies for diabetes, hypertension and hypercholesterolemia along the entire continuum of care (CoC)-from awareness and diagnosis, through treatment and control. In this study, we present baseline findings from HealthRise South Africa, identifying gaps in the CoC, as well as key barriers to care for non-communicable diseases (NCDs).

METHODS

This mixed-methods needs assessment utilized national household data, health facility surveys, focus group discussions, and key informant interviews in Umgungundlovu and Pixley ka Seme districts. Risk factor and disease prevalence were estimated from the South Africa National Health and Nutrition Examination Survey. Health facility surveys were conducted at 86 facilities, focusing on essential intervention, medications and standard treatment guidelines. Quantitative results are presented descriptively, and qualitative data was analyzed using a framework approach.

RESULTS

46.8% of the population in Umgungundlovu and 51.0% in Pixley ka Seme were hypertensive. Diabetes was present in 11.0% and 9.7% of the population in Umgungundlovu and Pixley ka Seme. Hypercholesterolemia was more common in Pixley ka Seme (17.3% vs. 11.1%). Women and those of Indian descent were more likely to have diabetes. More than half of the population was found to be overweight, and binge drinking, inactivity and smoking were all common. More than half of patients with hypertension were unaware of their disease status (51.6% in Pixley ka Seme and 51.3% in Umgungundlovu), while the largest gap in the diabetes CoC occurred between initiation of treatment and achieving disease control. Demand-side barriers included lack of transportation, concerns about confidentiality, perceived discrimination and long wait times. Supply-side barriers included limited availability of testing equipment, inadequate staffing, and pharmaceutical stock outs.

CONCLUSION

In this baseline assessment of two South African health districts we found high rates of undiagnosed hypercholesterolemia and hypertension, and poor control of hypercholesterolemia, hypertension, and diabetes. The HealthRise Initiative will need to address key supply- and demand-side barriers in an effort to improve important NCD outcomes.

The effect of quality of care on cardiovascular risk factors in newly diagnosed diabetic patients

BACKGROUND

In this study, we evaluated the quality of care and control of cardiovascular risk factors in newly diagnosed diabetic patients, identified during diabetes screening program, 1 year after diagnosis.

METHODS

In this prospective study, 83 newly diagnosed diabetic patients identified at screening in Isfahan, were studied. Height, weight, blood pressure, plasma glucose, lipids, and hemoglobin A1c (HbA1c) of these patients were measured 2 times, first at the time of diagnosis and then 1 year later, and the results were compared between two groups, with and without regular course of treatment.

RESULTS

Nearly 46.99% and 53.1% of the studied patients have regular and irregular course of treatment. After 1 year, significant improvement in the mean of plasma glucose, cholesterol, triglyceride, low density lipoprotein (LDL), high density lipoprotein and HbA1c was seen in patients with regular course of treatment except for blood pressure (P < 0.05). Frequency of controlled cardiovascular risk factors including fasting plasma glucose, HbA1c, cholesterol and LDL was significantly improved in patients with regular course of treatment (P < 0.05). Mentioned changes were not seen in patients with irregular course of treatment.

CONCLUSIONS

The findings of the current study demonstrated that though diabetes screening program result in earlier diagnosis of patients with type 2 diabetes, but it seems that regular follow-up and proper management of newly diagnosed patients is crucial for appropriate glycemic and metabolic control and preventing its related micro and macrovascular complication.

Systematic review and meta-analysis of validation studies on a diabetes case definition from health administrative records

OBJECTIVES

Health administrative data are frequently used for diabetes surveillance. We aimed to determine the sensitivity and specificity of a commonly-used diabetes case definition (two physician claims or one hospital discharge abstract record within a two-year period) and their potential effect on prevalence estimation.

METHODS

Following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, we searched Medline (from 1950) and Embase (from 1980) databases for validation studies through August 2012 (keywords: "diabetes mellitus"; "administrative databases"; "validation studies"). Reviewers abstracted data with standardized forms and assessed quality using Quality Assessment of Diagnostic Accuracy Studies (QUADAS) criteria. A generalized linear model approach to random-effects bivariate regression meta-analysis was used to pool sensitivity and specificity estimates. We applied correction factors derived from pooled sensitivity and specificity estimates to prevalence estimates from national surveillance reports and projected prevalence estimates over 10 years (to 2018).

RESULTS

The search strategy identified 1423 abstracts among which 11 studies were deemed relevant and reviewed; 6 of these reported sensitivity and specificity allowing pooling in a meta-analysis. Compared to surveys or medical records, sensitivity was 82.3% (95%CI 75.8, 87.4) and specificity was 97.9% (95%CI 96.5, 98.8). The diabetes case definition underestimated prevalence when it was ≤10.6% and overestimated prevalence otherwise.

CONCLUSION

The diabetes case definition examined misses up to one fifth of diabetes cases and wrongly identifies diabetes in approximately 2% of the population. This may be sufficiently sensitive and specific for surveillance purposes, in particular monitoring prevalence trends. Applying correction factors to adjust prevalence estimates from this definition may be helpful to increase accuracy of estimates.

Screening blood donors for diabetes: analysis of use, accuracy, and cost

BACKGROUND

The objective was to determine if a free, voluntary diabetes screening program as a part of the blood donation process might be cost-effective.

STUDY DESIGN AND METHODS

During the first 6 months of the program, 26,415 donors were screened using a single random plasma glucose (RPG) level. All donors were asked to eat before donation. Low-, moderate-, and high-risk groups were formed based on RPG levels (<140, 140-200, and >200 mg/dL). Contact with a telephone questionnaire was made with 139 of 178 (78%) of the persons in the high-risk group with 33 new cases of diabetes diagnosed by the donor's physician and 26 donors indicating that they were not diagnosed with diabetes. Sex- and age-matched donors in the low- and moderate-risk groups were contacted and administered the same questionnaire.

RESULTS

The three risk groups were similar, except for body mass index (28.1 ± 5.4 kg/m2 vs. 29.9 ± 5.5 kg/m2 vs. 32.7 ± 5.6 kg/m2 , p < 0.001). The discriminative effectiveness of screening was evaluated by the area under the receiver operating characteristics (AROC) curve. The AROC curve was 0.950 (95% confidence interval, 0.920-0.979) for the identification of diabetes. Using a RPG cutoff of 200 mg/dL, sensitivity was 100%, specificity was 82%, and positive predictive value was 56%. Cost analyses showed that the mean cost to screen, per donor, was less than $1. Cost per case identified was estimated to be less than $500 for a RPG cutoff of 200 mg/dL.

CONCLUSIONS

Screening during the blood donation process appears to be accurate, convenient, and inexpensive.

An opportunistic pre-diabetes screening program offered with existing hypertension screening

Diabetes mellitus refers to a group of metabolic diseases that share the hallmark characteristic of hyperglycemia. Generally, Diabetes is categorized as type I, or type II. Type I results from the body's failure to synthesize insulin, and requires insulin injections. Type II, also known as adult-onset or non-insulin dependent diabetes mellitus (NIDDM), occurs when the body's cells fail to use insulin properly due to a defective insulin receptor, and may also be combined with a relatively reduced insulin secretion. Type II can be managed with healthy lifestyle habits and early detection of high sugar levels. Most local health departments across New York State offer hypertension screening but no pre-diabetes screening programs. The US preventive Services Task Force recommends that asymptomatic adults with sustained blood pressure greater than 135/80 mm Hg should be screened for type II diabetes. Since high blood sugar levels can be controlled, and in some cases reduced, there exist strong benefits in offering pre-diabetic screening for individuals who are hypertensive.

The relative contribution of patient, provider and organizational influences to the appropriate diagnosis and management of diabetes mellitus

OBJECTIVE

To estimate the relative contribution of patient attributes, provider characteristics and organizational features of the doctors' workplace to the diagnosis and management of diabetes.

RESEARCH DESIGN AND METHODS

In a factorial experimental design doctors (n = 192) viewed clinically authentic vignettes of 'patients' presenting with identical signs and symptoms. Doctor subjects were primary care doctors stratified according to gender and level of experience. During an in-person interview scheduled between real patients, doctors were asked how they would diagnosis and manage the vignette 'patients' in clinical practice.

RESULTS

This study considered the relative contribution of patient, doctor and organizational factors. Taken together patient attributes explained only 4.4% of the variability in diabetes diagnosis. Doctor factors explained only 2.0%. The vast majority of the explained variance in diabetes diagnosis was due to organizational factors (14.3%). Relative contributions combined (patient, provider, organizational factors) explained only 20% of the total variance.

CONCLUSION

Attempts to reduce health care variations usually focus on the education/activation of patients, or increased training of doctors. Our findings suggest that shifting quality improvement efforts to the area which contributes most to the creation and amplification of variations (organizational influences) may produce better results in terms of reduced variations in health care associated with diabetes.

International challenges without borders: a descriptive study of family physicians' educational needs in the field of diabetes

BACKGROUND

The optimal care of persons with diabetes by general practitioners and family physicians (GP/FP) is complex and requires multiple competencies. This is a fairly unrecognized key challenge in the healthcare systems. In some cases, local and national Continuous Professional Development (CPD) initiatives target these challenges; however there have been few international initiatives, possibly because challenges emerging from different studies have not been linked across national boundaries. In this context, the authors have compiled data about gaps and/or barriers inherent to GP/FP care of persons with type 2 diabetes from Austria, Canada, Germany and the United Kingdom.

METHODS

Secondary analyzes of pre-existing studies were conducted to identify challenges in the care of patients with type 2 diabetes as faced by GPs/FPs. Two sources of data were reviewed: unpublished research data from collaborating organizations and articles from a literature search (in English and German). Articles retrieved were scanned by the research team for relevance to the study objectives and to extract existing gaps and barriers. The identified challenges were then categorized along three major axes: (1) phase of the continuum of care {from screening to management}; (2) learning domain {knowledge, skills, attitudes, behavior, context}; and (3) by country/region. Compilation and categorization were performed by qualitative researchers and discrepancies were resolved through discussion until concordance was achieved.

RESULTS AND DISCUSSION

Thirteen challenges faced by GPs/FPs in the care for patients with type 2 diabetes were common in at least 3 of the 4 targeted countries/regions. These issues were found across the entire continuum of care and included: pathophysiology of diabetes, diagnostic criteria, treatment targets assessment, drugs' modes of action, decision-making in therapies, treatment guidelines, insulin therapy, adherence, management of complications, lifestyle changes, team integration, bureaucracy and third-party payers. The issues reported were not restricted to the physicians' knowledge, but also related to their skills, attitudes, behaviours and context.

CONCLUSIONS

This study revealed challenges faced by GPs/FPs when caring for patients with diabetes, which were similar across international and health system borders. Common issues might be addressed more efficiently through international educational designs, adapted to each country's healthcare system, helping develop and maintain physicians' competencies.

Validity of the primary care diagnosis of diabetes in veterans in the southeastern United States

AIMS

To determine the validity of diagnosis of diabetes in primary care.

METHODS

Patients with initial primary care diagnosis (ICD-9 code 250.xx) were compared to matched controls (without code or diabetes drugs), and patients meeting VA Diabetes Epidemiology Cohort (DEpiC) criteria (any 250.xx twice, or diabetes drug) in "diagnostic accuracy" (whether hyperglycemia preceded diagnosis) and "predictive accuracy" (whether diabetes drug or A1c ≥ 6.5% followed diagnosis).

RESULTS

Only 1.8% of primary care diagnoses met ADA criteria, while nonstandard non-fasting morning glucose ≥ 126 mg/dl or A1c ≥ 6.5% were utilized in 51.5%; broad "diagnostic accuracy" criteria were met in 53% of 2980 primary care vs. 2% of 13,397 control (p<0.001), and 60% of 2456 DEpiC patients (p<0.001). "Predictive accuracy" was 88% in primary care diagnosis vs. 12% control (p<0.001) and 93% DEpiC patients (p=0.08), but was higher if ADA criteria were met. Delay from hyperglycemia to diagnosis averaged 12.5 months in primary care vs. 20.1 months in DEpiC patients (p<0.001).

CONCLUSIONS

While generally not based on ADA criteria, the primary care diagnosis of diabetes is valid, and identifies patients earlier than detection by DEpiC criteria. Either primary care diagnosis or DEpiC criteria could be used to trigger electronic reminders aimed to facilitate management.

The impact of clinical pharmacy services integrated into medical homes on diabetes-related clinical outcomes

BACKGROUND

Pharmacist services have expanded in the US health-care system from traditional roles to include comprehensive clinical services, but many studies lack comparison groups to evaluate outcomes of these clinical services.

OBJECTIVE

To evaluate the clinical outcomes of uninsured or underinsured patients with type 2 diabetes who received care from pharmacists in local "safety net" clinic medical homes compared to outcomes of patients from clinics receiving usual care without the services of clinical pharmacists.

METHODS

Pharmacists provided comprehensive pharmacy services in safety net clinic medical homes for uninsured patients in a major urban city. Referred patients had poor diabetes control (hemoglobin A(1c) [A1C] >9%). Pharmacists conducted comprehensive evaluations of medications, made adjustments, monitored adherence, and provided education and follow-up. Intervention patients were compared to similar patients who were receiving usual care but were not seen by a pharmacist. Outcomes evaluated were the change in A1C levels and achievement of treatment goals. Data were derived from chart reviews retrospectively. Multivariate least-squares and logistic regression models were used to estimate the impact of the intervention.

RESULTS

Two hundred twenty-two intervention and 262 control patients were evaluated. Patients receiving care from pharmacists had adjusted A1C levels reduced by 1.38% relative to usual care, increasing the likelihood of achieving an A1C <7% by 3-fold (p < 0.001 for both estimates).

CONCLUSIONS

The integration of clinical pharmacy services into safety net medical homes was associated with improvement in clinical outcomes of patients with diabetes.

The effect of major depression on preventive care and quality of life among adults with diabetes

OBJECTIVE

To examine the association between major depression and preventive care practices and quality-of-life indices among adults with diabetes.

METHODS

Data from 16,754 participants with diabetes in the 2006 Behavioral Risk Factor Surveillance Survey were examined. Multiple logistic regression was used to assess the independent association between the presence of major depression and a range of (1) preventive care behaviors and (2) quality-of-life indices, after accounting for relevant covariates.

RESULTS

With regard to preventive care practices, women with major depression were less likely to have received a mammogram in the past 2 years [OR 0.60 (95% CI 0.45-0.79) for women age 40 years and older and OR 0.52 (95% CI 0.37-0.71) for women age 50 and older] relative to their counterparts without major depression. With regard to quality-of-life indices, individuals with major depression were less likely to perceive their health as good [OR 0.24 (95% CI 0.19-0.31)], to report being satisfied with life [OR 0.10 (95% CI 0.07-0.14)] and to report receiving needed social support [OR 0.29 (95% CI 0.22-0.38)] relative to individuals without major depression. Individuals with major depression were also more likely to report one or more poor physical and poor mental health days in the past 30 days [OR 4.36 (95% CI 3.33-5.72) and OR 9.77 (95% CI 7.81-12.22), respectively] than those without major depression.

CONCLUSIONS

Major depression in adults with diabetes is associated with decreases in a wide range of quality-of-life indices as well as preventive self-care practices that were not entirely limited to diabetes-specific disease management.

The effect of depression on self-care behaviors and quality of care in a national sample of adults with diabetes

OBJECTIVE

To examine the effect of minor and major depression on self-care behaviors and quality of care among adults with diabetes.

METHODS

Data from 16,754 participants with diabetes in the 2006 Behavioral Risk Factor Surveillance Survey were examined. Multiple logistic regression was used to assess the independent association between depression status and indices of (1) self-care behaviors and (2) quality of diabetes care received, after accounting for confounders.

RESULTS

Individuals with minor (OR 0.69, 95% CI 0.57-0.84) and major (OR 0.50, 95% CI 0.39-0.64) depression were less likely to engage in leisure-time physical activity. Individuals with minor (OR 1.51, 95% CI 1.18-1.94) and major (OR 1.66, 95% CI 1.28-2.15) depression were more likely to be current smokers. With regard to quality of care, individuals with minor (OR 0.81, 95% CI 0.66-0.99) and major (OR 0.70, 95% CI 0.54-0.89) depression were less likely to receive an annual dilated eye exam. Additionally, individuals with minor (OR 0.79, 95% CI 0.65-0.95), but not major (OR 0.85, 95% CI 0.67-1.09) depression, were less likely to receive a flu shot in the past 12 months.

CONCLUSIONS

In adults with diabetes, both minor and major depression are associated with decreased self-care behavior and quality of care.

How are we diagnosing cardiometabolic risk in primary care settings?

AIMS

To assess attitudes, barriers and practices of clinicians in assessing and treating cardiometabolic risk in overweight adults.

METHODS

In 2006, primary care physicians and mid-level practitioners in Montana were surveyed (N=430).

RESULTS

Most primary care clinicians (95%) recognized the clinical benefit of weight loss, but many cited patient motivation (87%), lack of support services (61%), and lack of time (58%) as barriers. Over 80% identified obesity, hypertension, abnormal lipids, history of gestational diabetes, and family history as indications for diabetes screening. Most clinicians used fasting glucose (89%), random glucose (58%), and A1c (42%) as initial screens for diabetes. To confirm the diagnosis, the majority of respondents used A1c testing (80%) or fasting glucose (64%). Approximately one-quarter used the diagnosis pre-diabetes (26%), but just over half (52%) used alternative diagnoses of glucose intolerance. Sixty-five percent used the diagnosis of metabolic syndrome. Of those using metabolic syndrome, mid-level practitioners were more likely than physicians to assess waist circumference (49% vs. 63%).

CONCLUSIONS

Despite citing significant barriers, clinicians routinely assessed cardiometabolic risk with diabetes screening, but relatively few reported using the diagnosis pre-diabetes. Metabolic syndrome was used commonly to diagnose overweight adults at risk for diabetes and cardiovascular disease.

Trends in Method of Diagnosis of Type 2 Diabetes Mellitus: Results from SHIELD

Aims. This study assessed whether recent screening recommendations have led to increased diagnosis of type 2 diabetes mellitus (T2DM) through routine screening. Methods. Respondents to the 2006 US SHIELD survey reported whether a physician told them they had T2DM, age at diagnosis, specialty of the physician who made the diagnosis, and whether the diagnosis was made after having symptoms, during routine screening, or when being treated for another health problem. Results. Of 3 022 T2DM respondents, 36% of respondents reported that T2DM diagnosis was made during routine screening alone, 20% after having symptoms alone, and 6% when being treated for another health problem alone. The proportion of T2DM respondents reporting a diagnosis based only on screening increased approximately 42% over a 15+-year time span (absolute increase from 31% to 44%) (P < .001), whereas symptom-based diagnosis did not change significantly (P = .10). T2DM was diagnosed primarily by family physicians (88.3%). Conclusion. These findings highlight the importance of regular screening for diabetes and the vital role of primary care physicians in recognizing individuals with T2DM.

Quality of care for first-degree relatives of type 2 diabetes patients diagnosed with diabetes at a screening program one year after diagnosis

AIMS

Diabetes screening is an effective tool for diagnosing patients who are unaware of their diabetes and for providing them with optimal treatment. The quality of care and treatment of diabetic patients diagnosed at a screening program during one year in Isfahan, a centrally located Iranian city, was assessed.

METHODS

In a prospective study, 1640 first-degree relatives of diabetic patients (aged 35-55) were screened for diabetes mellitus at Isfahan Endocrine and Metabolism Research Center during 2003-2004. All patients diagnosed with diabetes during screening were selected and their height, weight, blood pressure, fasting plasma glucose, lipids and HbA1c were recorded at the time of diagnosis and one year later. The values at the time of screening were subsequently compared with those collected one year later.

RESULTS

Eighty-three subjects (5.06%) were diagnosed with diabetes during screening. Of these patients, 78.3% were dyslipidemic and 45% were hypertensive. One year after diagnosis, 77.1% of patients were receiving treatment for hyperglycemia. However, only 49.2% of dyslipidemic and 45% of hypertensive patients were being treated for these conditions. Body mass index, fasting plasma glucose, lipids and HbA1c had improved one year after diagnosis, but no significant improvement was observed in blood pressure.

CONCLUSIONS

The evaluation of the quality of care for newly diagnosed diabetic patients revealed that more attention should be paid to glycemic control and reducing cardiovascular risk factors, in particular hypertension.

The "trusted clinician": an alternative approach to worksite health promotion?

A trusted primary care clinician located in the workplace and serving all employees is presented as a desirable model for delivering prevention services to employees in this edition of The Art of Health Promotion. Many of the clinical studies that support this model are reviewed by the authors and their implications are discussed. Finally, the authors call for large scale experimentation to test the potential of the model to improve health risks status, lifestyle behaviors and health care utilization and cost.