Stephen B. Soumerai

Segmented regression analysis of interrupted time series studies in medication use research.

Interrupted time series design is the strongest, quasi‐experimental approach for evaluating longitudinal effects of interventions. Segmented regression analysis is a powerful statistical method for estimating intervention effects in interrupted time series studies. In this paper, we show how segmented regression analysis can be used to evaluate policy and educational interventions intended to improve the quality of medication use and/or contain costs.

Effects of Medicaid Drug-Payment Limits on Admission to Hospitals and Nursing Homes

BACKGROUND Many state Medicaid programs limit the number of reimbursable medications that a patient can receive. We hypothesized that such limitations may lead to exacerbations of illness or to admissions to institutions where there are no caps on drug reimbursements. METHODS We analyzed 36 months of Medicaid claims data from New Hampshire, which had a three-drug limit per patient for 11 of those months, and from New Jersey, which did not. The study patients in New Hampshire (n = 411) and a matched comparison cohort in New Jersey (n = 1375) were Medicaid recipients 60 years of age or older who in a base-line year had been taking three or more medications per month, including at least one maintenance drug for certain chronic diseases. Survival (defined as remaining in the community) and time-series analyses were conducted to determine the effect of the reimbursement cap on admissions to hospitals and nursing homes. RESULTS The base-line demographic characteristics of the cohorts were nearly identical. In New Hampshire, the 35 percent decline in the use of study drugs after the cap was applied was associated with an increase in rates of admission to nursing homes; no changes were observed in the comparison cohort (RR = 1.8; 95 percent confidence interval, 1.2 to 2.6). There was no significantly increased risk of hospitalization. Among the patients in New Hampshire who regularly took three or more study medications at base line, the relative risk of admission to a nursing home during the period of the cap was 2.2 (95 percent confidence interval, 1.2 to 4.1), and the risk of hospitalization was 1.2 (95 percent confidence interval, 0.8 to 1.6). When the cap was discontinued after 11 months, the use of medications returned nearly to base-line levels, and the excess risk of admission to a nursing home ceased. In general, the patients who were admitted to nursing homes did not return to the community. CONCLUSIONS Limiting reimbursement for effective drugs puts frail, low-income, elderly patients at increased risk of institutionalization in nursing homes and may increase Medicaid costs.

Effects of Limiting Medicaid Drug-Reimbursement Benefits on the Use of Psychotropic Agents and Acute Mental Health Services by Patients with Schizophrenia

BACKGROUND We examined the effects of a three-prescription monthly payment limit (cap) on the use of psychotropic drugs and acute mental health care by noninstitutionalized patients with schizophrenia. We hypothesized that reducing access to such drugs would increase the use of emergency mental health services and the rate of partial hospitalizations (full-day or half-day treatment programs) and psychiatric-hospital admissions. METHODS We linked Medicaid claims data for a period of 42 months with clinical records from two community mental health centers (CMHCs) and the single state psychiatric hospital in New Hampshire, where Medicaid imposed a three-prescription limit on reimbursement for drugs during 11 months (months 15 through 25) of the study. For comparison, we used Medicaid claims for a period of 42 months in New Jersey, which had no limit on drug reimbursement. The study patients (n = 268) and the comparison patients (n = 1959) were permanently disabled, noninstitutionalized patients with schizophrenia, 19 through 60 years of age, who were insured by Medicaid. We conducted interrupted time-series regression analyses to estimate the effects of the cap on the use of medications and mental health services. RESULTS The cap resulted in immediate reductions (range, 15 to 49 percent) in the use of antipsychotic drugs, antidepressants and lithium, and anxiolytic and hypnotic drugs (P < 0.01). It also resulted in coincident increases of one to two visits per patient per month in visits to CMHCs (range of increase, 43 to 57 percent; P < 0.001) and sharp increases in the use of emergency mental health services and partial hospitalization (1.2 to 1.4 episodes per patient per month), but no change in the frequency of hospital admissions. After the cap was discontinued, the use of medications and most mental health services reverted to base-line levels (measured in the first 14 months of the study). The estimated average increase in mental health care costs per patient during the cap (

Payment Restrictions for Prescription Drugs under Medicaid

1,530) exceeded the savings in drug costs to Medicaid by a factor of 17. CONCLUSIONS Limits on coverage for the costs of prescription drugs can increase the use of acute mental health services among low-income patients with chronic mental illnesses and increase costs to the government, even aside from the increases caused in pain and suffering on the part of patients.

Coverage by the News Media of the Benefits and Risks of Medications

Abstract In an attempt to contain costs, 27 Medicaid programs have implemented patient-level payment limits for medications, but the effects of these restrictions on quality of care, costs, and hea...

A Randomized Trial of a Program to Reduce the Use of Psychoactive Drugs in Nursing Homes

BACKGROUND The news media are an important source of information about new medical treatments, but there is concern that some coverage may be inaccurate and overly enthusiastic. METHODS We studied coverage by U.S. news media of the benefits and risks of three medications that are used to prevent major diseases. The medications were pravastatin, a cholesterol-lowering drug for the prevention of cardiovascular disease; alendronate, a bisphosphonate for the treatment and prevention of osteoporosis; and aspirin, which is used for the prevention of cardiovascular disease. We analyzed a systematic probability sample of 180 newspaper articles (60 for each drug) and 27 television reports that appeared between 1994 and 1998. RESULTS Of the 207 stories, 83 (40 percent) did not report benefits quantitatively. Of the 124 that did, 103 (83 percent) reported relative benefits only, 3 (2 percent) absolute benefits only, and 18 (15 percent) both absolute and relative benefits. Of the 207 stories, 98 (47 percent) mentioned potential harm to patients, and only 63 (30 percent) mentioned costs. Of the 170 stories citing an expert or a scientific study, 85 (50 percent) cited at least one expert or study with a financial tie to a manufacturer of the drug that had been disclosed in the scientific literature. These ties were disclosed in only 33 (39 percent) of the 85 stories. CONCLUSIONS News-media stories about medications may include inadequate or incomplete information about the benefits, risks, and costs of the drugs as well as the financial ties between study groups or experts and pharmaceutical manufacturers.

Adherence to National Guidelines for Drug Treatment of Suspected Acute Myocardial Infarction: Evidence for Undertreatment in Women and the Elderly

BACKGROUND Although psychoactive medications have substantial side effects in the elderly, these drugs are used frequently in nursing homes. Few interventions have succeeded in changing this situation, and little is known about the clinical effects of such interventions. METHODS We studied six matched pairs of nursing homes; at one randomly selected nursing home in each pair, physicians, nurses, and aides participated in an educational program in geriatric psychopharmacology. At base line we determined the type and quantity of drugs received by all residents (n = 823), and a blinded observer performed standardized clinical assessments of the residents who were taking psychoactive medications. After the five-month program, drug use and patient status were reassessed. RESULTS Scores on an index of psychoactive-drug use, measuring both the magnitude and the probable inappropriateness of medication use, declined significantly more in the nursing homes in which the program was carried out (experimental nursing homes) than in the control nursing homes (decrease, 27 percent vs. 8 percent; P = 0.02). The use of antipsychotic drugs was discontinued in more residents in the experimental nursing homes than in the control nursing homes (32 percent vs. 14 percent); the comparable figures for the discontinuation of long-acting benzodiazepines were 20 percent vs. 9 percent, and for antihistamine hypnotics, 45 percent vs. 21 percent. In the experimental nursing homes residents who were initially taking antipsychotic drugs showed less deterioration on several measures of cognitive function than similar residents in the control facilities, but they were more likely to report depression. Those who were initially taking benzodiazepines or antihistamine hypnotic agents reported less anxiety than controls but had more loss of memory. Most other measures of clinical status remained unchanged in both groups. CONCLUSIONS An educational program targeted to physicians, nurses, and aides can reduce the use of psychoactive drugs in nursing homes without adversely affecting the overall behavior and level of functioning of the residents.

Patients At-Risk for Cost-Related Medication Nonadherence: A Review of the Literature

BACKGROUND Evidence-based guidelines for the treatment of patients with acute myocardial infarction (AMI) have been published and disseminated by the American College of Cardiology and the American Heart Association. Few studies have examined the rates of adherence to these guidelines in eligible populations and the influence of age and gender on highly effective AMI treatments in community hospital settings. METHODS Medical records of 2409 individuals admitted to 37 Minnesota hospitals between October 1992 and July 1993 for AMI, suspected AMI, or rule-out AMI, and meeting electrocardiographic, laboratory, and clinical criteria suggestive of AMI were reviewed to determine the proportion of eligible patients who received thrombolytic, beta-blocker, aspirin, and lidocaine hydrochloride therapy. The effects of patient age, gender, and hospital teaching status on the use of these treatments were estimated using logistic regression models. RESULTS Eligibility for treatment ranged from 68% (n=1627) for aspirin therapy, 38% (n=906) for lidocaine therapy, and 30% (n=734) for thrombolytic therapy to 19% (n=447) for beta-blocker therapy. Seventy-two percent of patients eligible to receive a thrombolytic agent received this therapy; 53% received beta-blockers; 81% received aspirin; and 88% received lidocaine. Among patients ineligible for lidocaine therapy (n=1503), 20% received this agent. Use of study drugs was lower among eligible elderly patients, especially those older than 74 years (thrombolytic agent: odds ratio, 0.2; 95% confidence interval, 0.1 to 0.4; aspirin: odds ratio, 0.4, 95% confidence interval, 0.3 to 0.6; beta-blocker: odds ratio, 0.4; 95% confidence interval, 0.2 to 0.8). Female gender was associated with lower levels of aspirin use among eligible patients (odds ratio, 0.7; 95% confidence interval, 0.6 to 0.9); and there was a trend toward lower levels of beta-blocker and thrombolytic use among eligible women. CONCLUSIONS Use of lifesaving therapies for eligible patients with AMI is higher than previously reported, particularly for aspirin and thrombolytic use in nonelderly patients. Lidocaine is still used inappropriately in a substantial proportion of patients with AMI. Increased adherence to AMI treatment guidelines is required for elderly patients and women.

Cost-related medication nonadherence and spending on basic needs following implementation of Medicare Part D.

ObjectiveUp to 32% of older patients take less medication than prescribed to avoid costs, yet a comprehensive assessment of risk factors for cost-related nonadherence (CRN) is not available. This review examined the empirical literature to identify patient-, medication-, and provider-level factors that influence the relationship between medication adherence and medication costs.DesignWe conducted searches of four databases (MEDLINE, CINAHL, Sciences Citations Index Expanded, and EconLit) from 2001 to 2006 for English-language original studies. Articles were selected if the study included an explicit measure of CRN and reported results on covarying characteristics.Main ResultsWe found 19 studies with empirical support for concluding that certain patients may be susceptible to CRN: research has established consistent links between medication nonadherence due to costs and financial burden, but also to symptoms of depression and heavy disease burden. Only a handful of studies with limited statistical methods provided evidence on whether patients understand the health risks of CRN or to what extent clinicians influence patients to keep taking medications when faced with cost pressures. No relationship emerged between CRN and polypharmacy.ConclusionEfforts to reduce cost-related medication nonadherence would benefit from greater study of factors besides the presence of prescription drug coverage. Older patients with chronic diseases and mood disorders are at-risk for CRN even if enrolled in Medicare’s new drug benefit.

Effect of pay for performance on the management and outcomes of hypertension in the United Kingdom: interrupted time series study

CONTEXT Cost-related medication nonadherence (CRN) has been a persistent problem for individuals who are elderly and disabled in the United States. The impact of Medicare prescription drug coverage (Part D) on CRN is unknown. OBJECTIVE To estimate changes in CRN and forgoing basic needs to pay for drugs following Part D implementation. DESIGN, SETTING, AND PARTICIPANTS In a population-level study design, changes in study outcomes between 2005 and 2006 before and after Medicare Part D implementation were compared with historical changes between 2004 and 2005. The community-dwelling sample of the nationally representative Medicare Current Beneficiary Survey (unweighted unique n = 24,234; response rate, 72.3%) was used, and logistic regression analyses were controlled for demographic characteristics, health status, and historical trends. MAIN OUTCOME MEASURES Self-reports of CRN (skipping or reducing doses, not obtaining prescriptions) and spending less on basic needs to afford medicines. RESULTS The unadjusted, weighted prevalence of CRN was 15.2% in 2004, 14.1% in 2005, and 11.5% after Part D implementation in 2006. The prevalence of spending less on basic needs was 10.6% in 2004, 11.1% in 2005, and 7.6% in 2006. Adjusted analyses comparing 2006 with 2005 and controlling for historical changes (2005 vs 2004) demonstrated significant decreases in the odds of CRN (ratio of odds ratios [ORs], 0.85; 95% confidence interval [CI], 0.74-0.98; P = .03) and spending less on basic needs (ratio of ORs, 0.59; 95% CI, 0.48-0.72; P < .001). No significant changes in CRN were observed among beneficiaries with fair to poor health (ratio of ORs, 1.00; 95% CI, 0.82-1.21; P = .97), despite high baseline CRN prevalence for this group (22.2% in 2005) and significant decreases among beneficiaries with good to excellent health (ratio of ORs, 0.77; 95% CI, 0.63-0.95; P = .02). However, significant reductions in spending less on basic needs were observed in both groups (fair to poor health: ratio of ORs, 0.60; 95% CI, 0.47-0.75; P < .001; and good to excellent health: ratio of ORs, 0.57; 95% CI, 0.44-0.75; P < .001). CONCLUSIONS In this survey population, there was evidence for a small but significant overall decrease in CRN and forgoing basic needs following Part D implementation. However, no net decrease in CRN after Part D was observed among the sickest beneficiaries, who continued to experience higher rates of CRN.