Connie Mah

Insurance coverage of patient self-management devices like self-monitoring blood glucose (SMBG) equipment may help to reduce race-related barriers to effective care. We examined whether providing free home glucose monitors had greater impacts on self-monitoring among black versus white patients with diabetes. Using electronic medical record data (1992-1996), we used longitudinal survival analysis to examine racial differences in rates of initiation of SMBG after coverage and rates of discontinuation of SMBG 18 months after initiation. We used piecewise Cox models to compare relative rates of SMBG initiation between black and white patients before and after the policy. The study cohort included 2275 continuously enrolled adult patients with diabetes in a large, staff model HMO. Multivariate models were restricted to patients using oral therapy. Controlling for time-dependent and fixed effects, black patients were as likely to initiate SMBG as white patients before the policy (hazard ratio 1.14; 95% confidence interval 0.86-1.50) but more likely after the policy (hazard ratio 1.33; 95% confidence interval 1.01-1.76). Among postpolicy SMBG initiators, black patients were consistently at higher risk of SMBG discontinuation than white patients over time (P < 0.05). By the end of follow-up, discontinuation rates were 78% among black patients and 64% among white patients. The policy is effective in triggering additional diabetes patients to self-manage, particularly black patients. However, persistence after initiation of monitoring is short-lived. Although our results show the potential of such policies to narrow racial gaps in self-management among racial minority groups, further interventions may be needed to promote long-term adherence.

The use of lipid-lowering agents is suboptimal among dual enrollees, particularly blacks. To determine whether the removal of restrictive drug caps under Medicare Part D reduced racial differences among dual enrollees with diabetes. An interrupted time series with comparison series design (ITS) cohort study. A total of 8895 black and white diabetes patients aged 18 years and older drawn from a nationally representative sample of fee-for-service dual enrollees (January 2004-December 2007) in states with and without drug caps before Part D. We examined the monthly (1) proportion of patients with any use of lipid-lowering therapies; and (2) intensity of use. Stratification measures included age (less than 65, 65 y and older), race (white vs. black), and sex. At baseline, lipid-lowering drug use was higher in no drug cap states (drug cap: 54.0% vs. nondrug cap: 66.8%) and among whites versus blacks (drug cap: 58.5% vs. 44.9%, no drug cap: 68.4% vs. 61.9%). In strict drug cap states only, Part D was associated with an increase in the proportion with any use [nonelderly: +0.07 absolute percentage points (95% confidence interval, 0.06-0.09), P<0.001; elderly: +0.08 (0.06-0.10), P<0.001] regardless of race. However, we found no evidence of a change in the white-black gap in the proportion of users despite the removal of a significant financial barrier. Medicare Part D was associated with increased use of lipid-lowering drugs, but racial gaps persisted. Understanding non-coverage-related barriers is critical in maximizing the potential benefits of coverage expansions for disparities reduction.

Recent studies suggest that patients at greatest risk for diabetes complications are least likely to self-monitor blood glucose. However, these studies rely on self-reports of monitoring, an unreliable measure of actual behavior. The purpose of the current study was to examine the relationship between patient characteristics and self-monitoring in a large health maintenance organization (HMO) using test strips as objective measures of self-monitoring practice. This cross-sectional study included 4,565 continuously enrolled adult managed care patients in eastern Massachusetts with diabetes. Any self-monitoring was defined as filling at least one prescription for self-monitoring test strips during the study period (10/1/92-9/30/93). Regular SMBG among test strip users was defined as testing an average of once per day for those using insulin and every other day for those using oral sulfonylureas only. Measures of health status, demographic data, and neighborhood socioeconomic status we...

We sought to evaluate the effect of automated telephone assessment and self-care education calls with nurse follow-up on the management of diabetes. We enrolled 280 English- or Spanish-speaking adults with diabetes who were using hypoglycemic medications and who were treated in a county health care system. Patients were randomly assigned to usual care or to receive an intervention that consisted of usual care plus bi-weekly automated assessment and self-care education calls with telephone follow-up by a nurse educator. Outcomes measured at 12 months included survey-reported self-care, perceived glycemic control, and symptoms, as well as glycosylated hemoglobin (Hb A1c) and serum glucose levels. We collected follow-up data for 89% of enrollees (248 patients). Compared with usual care patients, intervention patients reported more frequent glucose monitoring, foot inspection, and weight monitoring, and fewer problems with medication adherence (all P -0.03). Follow-up Hb A,, levels were 0.3% lower in the intervention group (P = 0.1), and about twice as many intervention patients had Hb A1c levels within the normal range (P = 0.04). Serum glucose levels were 41 mg/dL lower among intervention patients than usual care patients (P = 0.002). Intervention patients also reported better glycemic control (P = 0.005) and fewer diabetic symptoms (P <0.0001 ), including fewer symptoms of hyperglycemia and hypoglycemia. Automated calls with telephone nurse follow-up may be an effective strategy for improving self-care behavior and glycemic control, and for decreasing symptoms among vulnerable patients with diabetes.

The objective of this study was to determine whether long-term benzodiazepine use is associated with dose escalation. The authors examined changes in dose and the frequency of dose escalation among new and continuing (at least two years) recipients of benzodiazepines identified from a database containing drug-dispensing and health care use data for all New Jersey Medicaid patients for 39 months. Independent variables included age; Medicaid eligibility category; gender; race or ethnicity; neighborhood socioeconomic variables; chronic illnesses, such as schizophrenia, bipolar illness, panic disorder, and seizure disorder; and predominant benzodiazepine received. Logistic regression analyses were conducted to determine the association between the independent variables and escalation to a high dosage (at least 20 diazepam milligram equivalents [DMEs] per day for elderly patients and at least 40 DMEs per day for younger patients). A total of 2,440 patients were identified, comprising 460 new and 1,980 continuing recipients. Seventy-one percent of continuing recipients had a permanent disability. Among all groups of continuing recipients, the median daily dosage remained constant at 10 DMEs during two years of continuous use. No clinically or statistically significant changes in dosage were observed over time. The incidence of escalation to a high dosage was 1.6 percent. Subgroups with a higher risk of dose escalation included antidepressant recipients and patients who filled duplicate prescriptions for benzodiazepines at different pharmacies within seven days. Elderly and disabled persons had a lower risk of dose escalation than younger patients. The results of this study did not support the hypothesis that long-term use of benzodiazepines frequently results in notable dose escalation.

Benzodiazepines (BZs) are safe, effective drugs for treating anxiety, sleep, bipolar, and convulsive disorders, but concern is often expressed about their overuse and potential for abuse. We evaluated the effects of physician surveillance through a Triplicate Prescription Program (TPP) on problematic and non-problematic BZ use. This study uses interrupted time series analyses of BZ use in the New York (intervention) and New Jersey (control) Medicaid programs for 12 months before and 24 months after the New York BZ TPP. The regulation required NY physicians to order BZs on triplicate prescription forms with one copy forwarded by pharmacies to a state surveillance unit. Study participants were community-dwelling persons over age 18 continuously enrolled between January 1988 and December 1990 in New York (n = 125,837) or New Jersey Medicaid (n = 139,405). During the baseline year, 20.2% of New York and 19.3% of New Jersey cohort members received at least one BZ prescription. After the TPP, there was a sudden, sustained reduction in BZ use of 54.8% (95% CI = [51.4%, 58.3%]) in New York with no changes in New Jersey. Significantly greater reductions were experienced by young women, and persons living in zip codes that were urban, predominantly Black, or with a high density of poor households. Increases in potential substitute medications were modest. At baseline, nearly 60% of BZ recipients had no evidence of potentially problematic use. Despite a somewhat greater likelihood of discontinuation of BZ therapy among those with potentially problematic use, the largest impact of the TPP was a substantially greater relative reduction in access to BZs among non-problematic users. State-mandated physician surveillance dramatically reduces BZ use with limited substitution of alternative drugs, lowers rates of possible abuse, but may severely limit non-problematic BZ use.

To examine the relationship between adherence to antidepressant medications and HbA1c levels among patients with diabetes in a managed care setting. The analysis included measures of HbA1c levels before, during, and after initial antidepressant use among 568 patients with diabetes enrolled in the Harvard Pilgrim Health Care insurance plan from 1991-1995. Adherence was defined as four refills in a six-month period after the first antidepressant prescription. General linear models using SAS PROC MIXED were used to estimate the effects of covariates including antidepressant adherence on HbA1c levels over time, comparing patients who were adherent to antidepressant medications to those patients who were non-adherent to antidepressant medications. Adherence to antidepressant treatment was not significantly associated with HbA1c levels among diabetic patients who are antidepressant users. Younger age, use of insulin and oral medications, and female gender were all significantly associated with HbA1c levels over time. Although we did not observe any association between level of adherence to antidepressant therapy among diabetic patients and levels of glucose control, our results confirm previously established associations between patient characteristics and glycemic control. Further research is needed to disentangle the complex relationship among antidepressant treatment adherence and diabetes outcomes.

To determine whether automated voice messaging (AVM) systems could be used as an adjunct to primary care for diabetic patients, we examined whether patients were able to respond to AVM queries for clinical information, whether sufficient numbers of problems were identified to warrant the implementation of the service, and whether patients found the system helpful. The AVM system we examined uses specialized computer technology to telephone patients, communicate messages, and collect information. Sixty-five diabetic patients participated. Based on a review of the literature and the input of diabetes clinician-researchers, we developed an AVM monitoring protocol to inquire about patients' symptoms, glucose monitoring, foot care, diet, and medication adherence. Patients also were given the option to listen to health promotion messages and to report their satisfaction with the calls. Patients responded by using their touch-tone telephone keypads. A total of 216 AVM calls were successfully completed, an average of 3.3 out of four calls per patient. Patients reported a variety of health problems that signaled the need for follow-up. Many patients reported not checking their blood glucose or their feet, and one in four reported problems with medication and diet adherence. Health and self-care problems varied across patient subgroups in ways suggesting that the AVM reports were reliable and valid. Overall, 98% of all patients reported that the calls were helpful, 98% reported that they had no difficulty responding to the calls, and 77% reported that receiving AVM calls would make them more satisfied with their health care. This study demonstrates that diabetic patients can respond to AVM queries and find the calls helpful. Such calls are a feasible strategy for identifying health and self-care problems that would otherwise go unnoticed by clinicians.