 To evaluate the effect of consumer cost sharing on use of physician-administered and patient self-administered specialty drugs for rheumatoid arthritis.

Multivariate statistical analysis of probability and use of physician-administered specialty drugs, patient self-injected specialty drugs, non-biologic disease-modifying anti-rheumaticdrugs, and symptom relief drugs. Analyses were conducted for patients enrolling in preferred provider organization (PPO) plans and health maintenance organization (HMO) plans with different cost-sharing requirements, adjusted for patient demographics, health status, and geographical location.

 Professional, facility, and pharmaceutical claims for beneficiaries of CalPERS, the public employee insurance purchasing alliance in California, for 2008-2009. Consumer cost-sharingrequirements were obtained for each type of drug and service for each type of insurance plan.

 PPO insurance enrollees face substantially higher cost sharing for physician-administered specialty drugs, compared with HMO enrollees in CalPERS. PPO patients with rheumatoidarthritis are only half as likely as HMO enrollees to choose a physician-administered specialty drug (4.2% vs 9.3%) (

.05), and use 25% less of the drugs if they use any ($10,356 vs $13,678) (

 <.05). They are 30% more likely to use a self-administered specialty drug than are HMO enrollees (29.3% vs 22.1%) (

.05), and use 35% more of the drugs if any ($16,015 vs $12,378) (

 Consumer cost sharing reduces the use of physician-administered specialty drugs for rheumatoid arthritis. The higher use of patient self-administered specialty drugs suggests that the disincentives for use of physician-administered drugs were offset by an increased incentive to use self-administered drugs.

		Consumer cost sharing influences the use of physician-infused and patient self-administered specialty drugs for complex diseases such as rheumatoid arthritis.

		Insurance benefit designs and cost sharing should be structured to encourage use of the most effective and cost-effective drugs, including high-cost biopharmaceuticals.

		Preferred provider organization insurance products tend to have higher coinsurance than health maintenance organization products, which is likely to influence patterns of use for office-administered drugs as well as for imaging tests and implantable devices.

Pharmaceutical innovation in recent years has come largely from biopharmaceuticals and other specialty drugs that need novel methods of handling, storage, transportation, and use.

 Many are administered to the patient by a physician in the office, in a community-based infusion center, or in a hospital ambulatory care center. Some new specialty drugs also can be administered  by the patient, often through self-injection. Physicians and patients with immunity disorders, cancer, and other complex conditions now often experiment with modes of administration to find the drug that best treats the patient’s symptoms and underlying disease.

The choice between office infusion and self-injection for biopharmaceuticals is based in part on clinical factors, as some patients respond better to one drug than to another despite the similarity in their reported effects and US Food and Drug Administration labeling.

 Convenience to the patient also plays an important role, with some patients preferring to avoid the time and cost of an office visit while others appreciate the single point of access, administration, and monitoring of side effects  at the office.

 Economic factors also may influence drug choices, to the extent the patient’s insurance coverage differentially requires cost sharing depending on the mode of administration.

The literature on the effects of consumer cost sharing on patient use and adherence to prescription drugs has been the subject  of 3 extensive reviews.

 All report that patients facing high deductibles, coinsurance, coverage limits, and cost sharing through tiered drug formularies are less likely to use prescribed drugs. Where outcomes data are available, the studies find worse reported outcomes among patients facing higher levels of cost sharing. However, only 1 of the published studies considers the impact of cost sharing on the use of physician office—administered specialty drugs and patient self-injected specialty drugs, given  the difficulty in obtaining data on specialty drug use and linking it to data on insurance benefit design and cost sharing. In that study,

 Karaca-Mandic and colleagues report that benefit generosity affected the likelihood that rheumatoid arthritis patients initiate and continue with specialty drug therapy. The authors did not find an impact of consumer cost sharing for physician-administered (medical benefit) as distinct from patient selfadministered (pharmacy benefit) specialty drugs.

This study evaluated the association between consumer cost sharing on the one hand and the choice between physician-administered and patient self-administered biopharmaceuticals for rheumatoid arthritis on the other. We analyze cost-sharing  requirements for patients under both the medical benefit (covering physician-administered drugs) and the pharmacy benefit (covering self-administered drugs). In addition to examining patterns of use and expenditure for biopharmaceuticals, it examined use and cost of symptom-relief medications and non-biologic disease-modifying anti-rheumatic drugs (DMARDs).

Data on Patients, Drugs, and Drug Expenditures

Comprehensive 2008-2009 medical insurance claims were obtained for employees, dependents, and retirees from the California Public Employees Retirement System (CalPERS), a multi-employer alliance that purchases health insurance coverage for the state of California and other public entities such as cities and school districts.

CalPERS professional, facility, and pharmaceutical claims were merged to create a file for each individual patient. The analysis was limited to individuals aged 18 to 64 years (as Medicare claims data were not available). Rheumatoid arthritis (RA) status was identified for an enrollee based on having experienced an episode of treatment for the condition, with episodes calculated from pharmacy, professional, and facility claims using the Thomson Reuters MedStat episode grouper.

The episode grouper is based on claims data for 43 million covered lives and uses expert opinion and the medical literature to develop criteria for selecting which of the claims filed on behalf of any 1 patient are associated with which underlying disease. Enrollees are identified as having particular underlying diseases based on the claims filed on their behalf by the treating physicians. Thomson Medstat is the manager of the CalPERS data warehouse.

Claims data were combined with demographic data from CalPERS on the individual’s age, gender, subscriber status (employee, dependent, retiree), and zip code of residence. Zip codes were used to assign individual members to 1 of 21 hospital referral regions, as defined by the Dartmouth Atlas as corresponding to local hospital markets.

 Since CalPERS was unable to provide information directly on each enrollee’s household income, we assigned to each patient the median household income in his or her zip code of residence, using data from the 2000 Census of Population.

Rheumatoid arthritis severity and presence of comorbidities were included using measures developed by Thomson Medstat. Thomson Medstat analyzed CalPERS claims to identify which patients had juvenile onset (as distinct from adult onset) RA and which had RA complications as defined by the presence of conduction disorder, pericarditis, cardiomyopathy, Felty’s syndrome, amyloidosis, anemia, pneumoconiosis, interstitial fibrosis, respiratory failure, uveitis, and/or vasculitides.

Comorbidities were identified in terms of whether the individual’s claims included 

 diagnosis codes for any of 12 broad categories of health conditions, including infectious and parasitic disease; noeoplasms; endocrine, nutritional, metabolic, and immunity disorders; diseases of the blood and blood-forming organs; mental disorders; diseases of the circulatory system; diseases of the respiratory system; diseases of the digestive system; diseases of the genitourinary system; diseases of the skin and subcutaneous tissue; and diseases of the musculoskeletal and connective tissue.

Expenditures were measured in terms of the total payment (allowed charge) for drugs used for the treatment of rheumatoid arthritis. The patient’s portion of the total drug payment (copayment and coinsurance requirements) is included as part of the allowed charge and measure of expenditure used in the study. Based on enrollment information, approximately 7% of enrollees in each of the 2 study years were enrolled for less than the full 12 months. We annualized their expenditure data by multiplying actual expenditures by the inverse of the fraction of the year for which they were enrolled. We also analyzed the data excluding these individuals.

Drugs were classified as to whether they were for relief of arthritic symptoms, DMARDs, or specialty drugs.

 Specialty drugs for this patient population include Orencia (abatacept), Rituxan (rituximab), Cimzia (certolizumab), Enbrel (etanercept), Humira (adalimumab), Kineret (anakinra), Remicade (infliximab), and Simponi (golimumab). DMARDs include methotrexate, leflunomide, sulphasalazine, gold sodium thiomalate, auranofin, antimalarials (hydroxychloroquine, chloroquine phosphate, chloroquine sulphate), minocycline, penicillamine, cyclophosphamide, cyclosporine, azathioprine, misoprostol, methylprednisolone, and aurothioglucose.

Symptom relief drugs include analgesics, non-steroidal antiinflammatory drugs, and corticosteroids.

Patient self-administered drugs were obtained from the CalPERS pharmaceutical claims data. Drugs administered by physicians in an office setting were obtained from the professional and facility claims data. Four expenditure measures were calculated: annual expenditures on physician-administered specialty drugs, patient self-administered specialty drugs, DMARDs, and symptom relief drugs.

Data on Insurance and Consumer Cost Sharing

. CalPERS members are offered a choice among 3 principal forms of health insurance coverage. Individuals may select KaiserPermanente, a large group model health maintenance organization (HMO); Blue Shield of California, which manages medical groups and Independent Practice Associations for the CalPERS network model HMO; or Anthem Blue Cross, which contracts with physician practices for the CalPERS preferred provider organization (PPO) product. This study is limited to CalPERS members selecting either Blue Shield HMO or Anthem PPO, since the Kaiser Permanente system does not generate claims data comparable to those generated by the other plans.

CalPERS has standardized some elements of the insurance benefit package and consumer cost-sharing requirements for PPO and HMO plans, while leaving others distinct. Costsharing requirements differ between the PPO and HMO for the medical benefit, which covers professional and facility claims plus claims for physician-administered drugs.

PPO members face a $500 annual deductible (individuals) and $1000 deductible (family) under the medical benefit. PPO patients must pay 20% of the cost of office-administered specialty products such as biologics and allergy treatments. Physician office visits for PPO patients require a $20 copayment if done in-network and 40% coinsurance if done out of network. HMO patients must pay $15 for a physician visit but pay nothing for physician-administered products. HMO member cost sharing is limited to a maximum of $1500 per year ($3000 for families). PPO members face a higher costsharing maximum ($2000-3000 for individuals, $4000-$6000 for families).

Cost-sharing requirements under the pharmacy benefit, which covers patient self-administered drugs obtained from a retail pharmacy, mail order pharmacy, or specialty pharmacy, have been standardized by CalPERS and are similar, though not identical, for the PPO and HMO plans. All patients must pay $5 for generic drugs, $10 for preferred brand drugs, and $45 for non-preferred brand drugs for the first month’s supply. PPO patients must pay $10, $25, and $75 for these drugs after the first month’s supply. Self-administered drug cost-sharing payments accrue to the patient’s pharmacy benefit maximum while physician-administered drug cost-sharing payments accrue to the patient’s medical benefit maximum. Pharmacy benefit cost sharing for both PPO and HMO enrollees is capped at $1000 per year.

Expenditure differences for self-administered drugs reflect differences between PPO and HMO patients in the volume of use rather than in drug prices, because self-administered drugs are purchased for both PPO and HMO enrollees by the CalPERS-contracted pharmacy benefit management firm. However, expenditure differences for physician-administered drugs may reflect price as well as volume differences if the Anthem PPO and Blue Shield HMO reimburse physician practices at different rates for the same physician-infused specialty drugs. Provider payment rates are confidential, but the CalPERS staff expressed confidence that prices for physician-administered  drugs were similar across the PPO and HMO plans.

In summary, the PPO and HMO enrollees face costsharing requirements that differ in several dimensions, most importantly with respect to physician-administered specialty drugs. In the statistical analysis, we use plan type (PPO vs HMO) as an indicator of the mix of cost-sharing requirements facing each patient, with PPO enrollment requiring substantially more cost sharing than HMO enrollment for physicianadministered drugs but similar cost-sharing requirements for self-administered drugs. We expect to observe that the higher cost-sharing requirement for physician-administered drugs but similar cost-sharing requirement for self-administered  drugs will induce PPO enrollees to use less of the first category and more of the second category of drugs  compared with HMO enrollees.

 We calculated mean expenditures on each category of drug (physician-administered specialty drugs, patient self-administered specialty drugs, non-biologic DMARDs, symptom relief drugs) and compared differences for PPO and HMO enrollees. It is possible that PPO and HMO enrollees differ in drug expenditures due to variance in demographic, health status, or geographic characteristics. We therefore computed adjusted mean expenditures after taking into account patient-specific characteristics.

While many RA patients use symptom-relief drugs and DMARDs, most do not use specialty drugs in any given year. We therefore employed a 2-part model to examine the association between type of insurance coverage and the probability of use of each type of specialty drug (office-administered and self-administered), adjusting for demographic characteristics, disease severity, comorbidities, and geographic location.

This statistical approach first analyzes the decision to use a physician-administered or patient self-administered specialty drug (eg, to have zero or positive expenditures on each of these classes of drugs). The model then analyzes expenditures during the year for patients who use any of each type of specialty drug (eg, have nonzero expenditures). The model is implemented through a probit equation (whether the patient uses any drug or not), and a generalized least-squares regression equation (annual expenditures, for patients with nonzero expenditures).

We use 2 years of data (2008 and 2009) because RA is a chronic condition subject to periodic flare-ups in symptoms and hence in patient requests for treatment. The same patient may have large drug-related expenditures in one year but low expenditures in the next. Including 2 years of data on most patients (this is an unbalanced panel) allows the analysis to better pick up the association between cost sharing and drug use, since its smoothes out some of the fluctuations in disease severity. In order to account for the presence of some patients in both years’ data, standard errors in the statistical analyses incorporate clustering at the patient level.18 In order to check on the impact of combining 2 years’ data (and thereby counting the same individual twice), we also conducted analyses for each of the 2 years separately. A total of 3445 CalPERS enrollees with  rheumatoid arthritis appear in the data: 2687 appear in both 2008 and 2009, 69 are present only in 2008, and 111 are present only in 2009. The analytic data set was constructed as a 2-year panel creating a total analytic sample of 5554.

We use the output of the 2-part statistical model to calculate adjusted mean expenditures for PPO and HMO patients, controlling for demographic, health, and geographic variables. We present adjusted mean expenditures for the full sample of PPO and HMO patients, the probability of use of physicianadministered and patient self-administered drugs for each group, and mean expenditures for those who have used each type of drug. These adjusted figures can be interpreted as the expenditures and probabilities of use that would be observed if PPO and HMO enrollees had the same demographic, health, and geographic characteristics.

 presents descriptive statistics on CalPERS members suffering from rheumatoid arthritis, both for the full sample and for those selecting the PPO and HMO plans, respectively. The majority (78%) of patients are women, consistent with the higher RA disease prevalence among women in the general population.19,20 Small minorities of CalPERS enrollees have juvenile onset RA (1.5%) or RA with complications (3.5%).

Total drug expenditures for treatment of rheumatoid arthritis average $4975 per year and are 12% higher for PPO enrollees ($5572) than for HMO enrollees ($4449). Total drug expenditures are categorized into physician-administered specialty drugs, patient self-administered specialty drugs, DMARDs, and symptom relief drugs for the treatment of RA. Three-fourths of expenditures are incurred for patient self-administered specialty drugs, with most of the remainder incurred for physician-administered drugs. Use of non-biologic DMARDs and symptom-relief drugs is quite common but is inexpensive relative to the use of specialty drugs.

The choice between physician-administered and self-administered drugs is made quite differently by PPO and HMO enrollees. While 85% of specialty drug use by PPO enrollees is for self-administered drugs ($4730 per patient per year) and only 9% is for physician-administered drugs ($498), 64% of drug use by HMO enrollees is for self-administered specialty drugs ($2828) and 32% is for physician-administered drugs ($1428). These figures refer to expenditures averaged across the entire PPO and HMO enrollee populations with rheumarheumatoid arthritis, not merely across those enrollees who use these drugs in the given year.

As indicated in the second section of Table 1, the percent using self-administered specialty drugs (26.0%) is much higher than the percent using physician office—administered drugs (7.8%). However, the ratio of self-to-office administered drugs is much greater for PPO enrollees (30.0% to 4.5%) than for HMO enrollees (22.9% to 10.7%). Annual expenditures are much higher for patients using self-administered drugs ($3720) than for patients using office-administered drugs ($991).

 presents multivariate regression parameters for the association between type of insurance coverage (PPO vs HMO), on the one hand, and use of and annual expenditure on specialty drugs, on the other, after adjusting for patient characteristics, health status, and geographic location. The first and third columns of the table present the marginal effect of the covariates on the probability of use, derived from maximum likelihood probit estimates. Adjusting for patient characteristics, PPO enrollees are 5.5 percentage points (

.01) less likely to use physician-administered specialty drugs and 7.1 percentage points (P <.01) more likely to use patient selfadministered specialty drugs than are HMO enrollees. These differences are to be evaluated against the baseline use rates of 7.8% for physician-administered and 26.0% for patient selfadministered specialty drugs in the CalPERS population suffering from rheumatoid arthritis.

Age, gender, and household income are not significantly associated with use of specialty drugs. Patients with juvenile onset RA  are 14.5 percentage points (

.01) more likely to use self-administered specialty drugs than are those with adult onset RA. Patients with more severe RA are 5.1 percentage points (

.01) more likely to use physicianadministered specialty drugs than are those with less severe RA conditions. Employment status, comorbidities, and geographic location are not consistently associated with use of specialty drugs.

The second and fourth columns of Table 2 present generalized least squares parameter estimates for the association between type of insurance plan and annual patient expenditures on specialty drugs. These analyses are limited to patients incurring positive expenditures on these drugs during the year. After adjusting for patient demographic, health status, and geographic characteristics, PPO enrollees who use physician-administered specialty drugs incur annual expenditures $3322 lower (

.01) than do HMO enrollees who use physician-administered drugs. PPO enrollees who use self-administered drugs incur annual expenditures $3637 higher (

.01) than do HMO enrollees who use self-administered specialty drugs.

 presents the adjusted probabilities of use for physician-administered and patient self-administered specialty drugs, and annual expenditures on those drugs, for PPO and HMO  enrollees. They are computed based on the regression parameters reported in Table 2. PPO patients are only half as likely to choose a physician-administered specialty drug as are HMO enrollees (4.2% vs 9.3%) (

.05). They are 30% more likely to use a self-administered specialty drug than are HMO enrollees (29.3% vs 22.1%) (

.05), and use 35% more of the drugs if they use any ($16,015 vs $12,378) (

Patients suffering from rheumatoid arthritis incur high medical care expenditures, with the majority of expenditures accounted for by physician-infused and patient self-administered specialty drugs. The choice of treatment and mode of administration are based in large part on clinical evidence on effectiveness and the patient’s experience. However, economic incentives stemming from the insurance benefit design and cost-sharing requirements also are strongly associated with these choices.

This study compared the use and cost of specialty drugs for rheumatoid arthritis patients enrolled in PPO and HMO insurance plans, respectively. The benefit designs and costsharing requirements across these 2 insurance types were standardized in part by the multi-employer purchasing alliance that provides the benefits, but also varied in meaningful ways. The principal difference was a higher overall level of cost sharing required of PPO enrollees seeking to access physician-administered drugs.

The study found that differential cost-sharing requirements were significantly associated with differences in the probability of use and expenditures on physician-administered and selfadministered specialty drugs. After adjusting for patient demographic, health status, and geographic residence differences,PPO enrollees were less likely to use physician-administered specialty drugs and more likely to use patient self-administered specialty drugs than were HMO enrollees. PPO enrollees who used physician-administered drugs incurred lower annual expenditures on those drugs, indicating less frequent use and/or lower doses, than did HMO enrollees. In contrast, PPO enrollees who used self-administered drugs incurred higher annual expenditures for those drugs, indicating more frequent use and/or higher doses, compared with HMO enrollees.

Articles

WEB EXCLUSIVE

Consumer cost sharing reduces use of physician-administered specialty drugs for patients suffering from rheumatoid arthritis, with an offsetting increase in patient self-administered specialty drugs.

Consumer Cost Sharing and Use of Biopharmaceuticals for Rheumatoid Arthritis

 To evaluate the effects of the Medicare Part D coverage gap on pharmacy use among a national sample of Medicare beneficiaries and on medication adherence among 2 subsamples with heart failure and/or diabetes.  

Pre-post design, with comparison group and propensity score weighting.  

We used a 5% random sample of elderly Medicare beneficiaries enrolled in stand-alone Part D plans in 2007. The comparison group had full coverage in the gap, whereas the 2 study groups had either no coverage or generic-only coverage in the gap. Main outcomes included probability of filling a prescription, monthly pharmacy spending and number of prescriptions filled, and adherence measured by medication possession ratios.

Relative to the comparison group, beneficiaries without drug coverage in the gap reduced the number of prescriptions filled per month by 16.0% (95% confidence interval [CI], 15.5%-16.5%); those with generic drug coverage in the gap reduced it by 10.8% (95% CI, 10.3%-11.4%). Most of the reduction was attributable to reduced use of brand-name drugs. Beneficiaries with heart failure reduced adherence to heart failure drugs by 3.6% (95% CI, 2.9%-4.2%) and beneficiaries with diabetes reduced antidiabetic medication adherence by 10.3% (95% CI, 9.4%-11.3%).  

Medicare beneficiaries reduced medication use (mainly brand-name drugs) after entering the coverage gap. This result suggests that while beneficiaries’ financial burden would continue because of the coverage gap, the gap would not result in a large reduction in medication adherence for essential drugs for diabetes and heart failure.  

 We examined the effect of the coverage gap on a national sample of Medicare beneficiaries as well as those with heart failure and/or diabetes. Although beneficiaries’ financial burden in the coverage gap will continue until 2020, the gap will not result in a large reduction in medication adherence for essential drugs.

		Medicare beneficiaries reduced medication use (mainly brand-name drugs) after entering the coverage gap.

		Beneficiaries with generic drug coverage in the gap reduced their drug use less.

		However, medication adherence was reduced to a much smaller degree for essential drugs used to treat heart failure and diabetes.

Medicare Part D offers prescription drug coverage for Medicare beneficiaries. Since the program’s inception in 2006, many beneficiaries have obtained improved drug coverage and increased their use of medications.

 The standard Part D benefit in 2007 included an initial $265 deductible, a period in which beneficiaries paid 25% of drug costs between $265 and $2400, a coverage gap in which they paid 100% of costs until their total out-of-pocket spending reached a catastrophic limit of $3850, and a catastrophic coverage period in which they paid 5%.

A major concern is the large coverage gap in the standard Part D design. About one-third of all Medicare beneficiaries enter this coverage gap each year.

 Faced with a sudden increase in medication costs, beneficiaries might cut back on their medication use and reduce medication adherence for the essential drugs.

That, in turn, might put their health at risk, leading to subsequent increases in hospitalizations and medical spending.

 Because of the high cost of Medicare and the potential implications for health outcomes, it is important to determine whether the coverage gap leads to reduction in medication adherence for essential drugs.

A few studies have examined the effects of the coverage gap on medication use, but the majority of these studies used pharmacy data either from 1 local Medicare-Advantage Part D plan

 One recent study used Medicare Part D data to examine the effects of the coverage gap among 2 subsamples with hypertension and hyperlipidemia.

 They found that the coverage gap wasassociated with reduction in use of and adherence to medications for hypertension and hyperlipidemia. Another recent study looked at aged Medicare beneficiaries with cardiovascular disease enrolled in a standalone Part D plan (PDP) or a retiree drug plan that was administered by CVS Caremark.

 They found that beneficiaries with no financial assistance during the coverage gap were more likely to discontinue their cardiovascular drugs.

In this study, we evaluated the effects of the coverage gap on medication spending and use among aged Medicare beneficiaries (those 65 years or older) in the national Part D data. In addition, we selected 2 subsamples of those with heart failure and/or diabetes to evaluate the gap effect on medication spending and adherence. We selected these 2 conditions because they are common in Medicare beneficiaries and because the medication treatment is very important— premature discontinuation may lead to increases in subsequent spending on medical care.

We determined how beneficiaries with these chronic conditions responded to the coverage gap compared with the general Medicare population.

Our analyses were guided by a number of hypotheses: (1) Medicare beneficiaries will decrease all their medication use regardless of drug classes once they enter the gap— and that the decrease will be higher for those with no drug coverage in the gap than for those with generic coverage in the gap.

However, those who anticipate going through the gap and entering the catastrophic period will not reduce their medication use in the gap.

 (2) Beneficiaries will decrease their use of all brand-name drugs more than generic drugs, and if beneficiaries have generic drug coverage they will shift from brand-name to generic drugs.

 (3) Among those with heart failure and/or diabetes, adherence to cardiovascular and diabetic drugs will decrease in the coverage gap.

We obtained beneficiary demographic and enrollment information, plan benefits, prescription drug events, and medical claims for a 5% random sample of all aged Medicare beneficiaries who were continuously enrolled in PDPs in 2007. We identified 2 subpopulations of beneficiaries: those with heart failure and/or diabetes. Each subpopulation was identified by 2 criteria: (1) having a diagnosis for the condition before January 1, 2007, defined by 2006 Centers for Medicare & Medicaid Services Chronic Condition Warehouse indicators (

; and (2) use of at least 1 medication for the condition in the initial coverage period so we could ensure that beneficiaries were on medications of interest before entering the coverage gap (

  provides the list of medications). The study design was approved by the institutional review board at the University of Pittsburgh.

Although the standard Part D benefit includes the 4 phases described in the first paragraph of this study, some companies offering PDPs modified the design and offered either “actuarially equivalent” or enhanced plans. In 2007, 72% of stand-alone PDPs had the standard coverage gap, 27% of PDPs had some generic coverage in the gap, and fewer than 1% offered coverage for both brand-name and generic drugs.

 In addition, beneficiaries with low incomes who are eligible for subsidies had a lower copayment throughout the year and did not face a coverage gap.

Using this variation in drug coverage, we identified 2 study groups based on the type of coverage they had in the gap (no coverage, generic only) and 1 comparison group. The comparison group consisted of beneficiaries who had either 12-month Medicaid coverage or 12-month low-income subsidies (LIS group). Thus, the drug coverage for the comparison group did not change throughout the year; that is, they did not face the gap in coverage. Even though the LIS group was different from the 2 study groups in socioeconomic characteristics, they could still serve the purpose of controlling for the underlying time trends in medication use, because their drug coverage did not change while the study groups had a sudden increase in drug costs once entering the coverage gap.

We used a pre-post study design with a comparison group to assess the impact of the coverage gap on medication and medical care use. This approach uses a difference-in-difference estimate, comparing the pregap and within-gap change for each study group with the pregap and within-gap change in the comparison group. This design does not require that study and comparison groups have the same baseline characteristics. As long as different groups have similar baseline trends, we could get unbiased results.

 We tested the baseline pregap trends in overall medication use across 3 groups, and they were similar.

We hypothesized that beneficiaries who went through the coverage gap and entered the catastrophic coverage period would not reduce their medication use in the gap.

 We ran regressions to test this hypothesis and found that those entering the catastrophic period did not change their medication use in the gap (

). Thus, for this study we primarily focused on beneficiaries who entered the gap but did not go through it. That is, each individual in our study sample had a pregap and a within-gap period.

To define pregap and within-gap periods, we first identified the index date as the first day that the beneficiary’s total drug spending reached the coverage gap threshold. The pregap period was defined as January 1, 2007, to the index date. The within-gap period was defined as the time from the first day after the index date until December 31, 2007. The index date was included in the pregap period because almost all prescriptions filled by beneficiaries on that day were subject to copayment levels in the initial coverage period. We compared the change in each outcome in the pregap and withingap periods between each study group and the comparison group: no coverage versus LIS and generic only versus LIS. We used a propensity score weighting mechanism to balance each study group with the comparison group. The propensity score weighting was conducted separately for each group: the general population and each subgroup of those with heart failure and/or diabetes.

For the pregap and within-gap periods separately, we defined 4 outcomes to measure medication use: (1) probability of using a medication (1 = used a medication; 0 = did not use a medication in the study period); (2) the mean number of monthly prescriptions filled per month, defined as the total number of prescriptions standardized by 30-day supply (ie, a prescription with a 90-day supply would count as 3 monthly drugs) divided by the number of months in the study period; (3) mean monthly pharmacy spending per month; and (4) medication adherence measured by medication possession ratio (MPR). For the general population, we focused on the first 3 outcomes. We measured these outcomes for all medications and then for brand-name and generic drugs separately.

For the patients with chronic conditions, we focused on the mean number of monthly prescriptions filled and the MPR. We examined the mean number of monthly prescriptions filled overall as well as the number of monthly prescriptions for disease-specific drugs. We defined the MPR as the proportion of days during a given period (eg, either the pregap or the within-gap period) that a subject had possession of any drugs used to treat the chronic illness. The prescriptions filled on the day the patient entered the coverage gap were included in calculating the within-gap MPRs for 2 reasons: (1) because these drugs were used in the within-gap period and (2) because inclusion of these drugs was consistent with the pregap MPR definition where drugs filled on the first day (January 1, 2007) were included in the calculation. Similarly, to be consistent with the MPR calculation in the pregap period where prescriptions filled before January 1, 2007, were not included, prescriptions filled before entering the gap were not included in the calculation for the withingap adherence. However, the definitions for MPRs should not change the difference-in-difference results because we applied the same rules for the study and comparison groups.

To implement the propensity score weighting mechanisms, we conducted a 2-stage analysis. In the first stage, we ran 2 logistic regression models to predict the probability of being in a study group relative to the comparison group, controlling for age, sex, race, number of Elixhauser comorbidities,

 and Prescription Drug Hierarchical Condition Category, the beneficiary risk adjuster used by Centers for Medicare & Medicaid Services to adjust payment to plans for pharmacy costs.

In the second stage, we ran a difference-in-difference model with the inverse of the propensity score as a weight. This effectively assigned a higher weight to individuals in the comparison group who had characteristics similar to those of individuals in the study group. In this model, the dependentvariable was the difference between within-gap and pregap periods for each previously defined outcome. Because pregap and within-gap measures were likely to be correlated, the advantage of this approach versus using 2 interrelated outcomes is that we could simply eliminate the correlated structure in 2 outcomes. The key independent variable was the indicator for being in the study group relative to the comparison group. All the covariates used in calculating propensity scores were included in the model.

In addition, we controlled for duration of time spent in the coverage gap (

) in the model because the longer beneficiaries stayed in the gap, the more likely it was that they would change their medication use and medical spending. All analyses were conducted using SAS software, version 9.2 (SAS Institute Inc, Cary, North Carolina) and R: A Language and Environment for Statistical Computing, version 2.12 (http://www.r-project.org).

 compares the characteristics between each study group and comparison group for beneficiaries whose pharmacy spending exceeded the coverage gap threshold but did not exceed the catastrophic coverage threshold. All the numbers are after adjustment with propensity score weights. After adjustment, all characteristics were comparable (ie, there was no statistically significant difference at 

 >.05) between each study group and the comparison group.  

Effects of the Coverage Gap on Medication Use and 

presents the effects of the coverage gap on the probability of using a drug, the mean number of monthly prescriptions filled, and the mean monthly pharmacy spending for all medications for the overall population. There are 3 main findings:

First, relative to the comparison group, there were statistically significant reductions in all of the outcomes (probability of using a drug, mean number of monthly prescriptions filled, and monthly pharmacy spending) in both study groups. However, those with no coverage generally decreased their use of medications more than those with generic drug coverage in the gap.

Second, the overall decrease in monthly medications and spending on drugs was primarily due to the decrease in the use of brand-name drugs. For example, those without drug coverage in the gap reduced their overall medication use by 0.85 medication per month (95% confidence interval [CI], 0.82-0.88); 75% of the reduction was accounted for by brandname drugs and 25% by generic drugs. This group decreased its monthly pharmacy spending by $73.15 (95% CI, $71.66- $74.65), of which $66.65 (95% CI, $65.30-$68.01) was for brand-name drugs and $6.40 (95% CI, $5.88-$6.92) was for generic drugs.  

Third, those with only generic coverage in the gap reduced their use of brand-name drugs but did not increase their use of generic drugs. In fact, they decreased their use of generic drugs slightly but negligibly. Among the general population, relative to the comparison group, the generic-only group reduced the number of monthly prescriptions filled by 0.66 (95% CI, 0.63-0.70); almost all of this decrease was attributable to the reduction in brand-name drugs (0.61 [95% CI, 0.59-0.62]).  

Effects of the Coverage Gap Among Patients With Chronic Conditions

Beneficiaries with heart failure and/or diabetes decreased their overall use of medications, and the overall decrease was similar to that found for the general population (

). They also decreased their use of the drugs specific to their conditions. The relative decreases in the condition-specific drugs were similar to those observed for the drugs overall.  

The decrease in medication use was accompanied by a decrease in medication adherence in both groups, but the decrease in adherence was smaller than the reduction in the number of prescriptions filled (

). In addition, relative to the comparison group, the decrease in medication adherence was always larger in the no-coverage group than in the generic-only group. In the heart failure sample, the MPR for heart failure drugs among the no-coverage group dropped from 0.87 in the pregap period to 0.83 in the within-gap period, while the MPR for those with generic-only coverage dropped from 0.87 to 0.86. Although statistically significant, the relative reduction in adherence for heart failure was negligible. The same pattern was observed for those with diabetes, although decrease in medication adherence was larger in this group.  

Medicare beneficiaries decreased their medication use once they entered the gap—and the decrease in use was consistently higher for those who had no coverage in the gap than for those with generic drug coverage in the gap. We hypothesized that beneficiaries would reduce their use of brand-name drugs after entering the coverage gap and if they had generic drug coverage, they would shift their brand-name drugs to generic drugs. We did observe that beneficiaries reduced their use of brand-name drugs substantially more than their use of generic drugs, but those with generic drug coverage did not switch from brand-name to generic drugs. This is partially because those in the generic-only group tended to use more generic drugs in the initial coverage gap period.  

We adjusted the time spent in the gap in our model so the beneficiaries spending less time in the gap contributed less to the regression results. In addition, we conducted sensitivity analysis by excluding those beneficiaries who stayed in the gap for less than 1 month or less than 2 months. The results were robust.  

Entering the Medicare Part D coverage gap does not result in a large reduction in medication adherence for essential drugs.

Effects of Medicare Part D Coverage Gap on Medication Adherence

 With the impetus for healthcare reform and the imperative for healthcare organizations to improve efficiency and reduce waste, it is valuable to examine high-volume procedures  and practices in order to identify potential overuse. At the same time, organizations must ensure that improved efficiency does not inadvertently reducepatient safety.

 We undertook a multicenter analysis of the use of adult cardiac telemetry outside of the intensive care unit or  step-down units at 4 teaching hospitals to determine the percentage of monitoring days that were not justified by an accepted indication and the monetary costs associated with these nonindicated days. We also assessed the safety of eliminating monitoring on days when it was not justified by looking at the incidence of arrhythmias.

We found that in 35% of telemetry days, telemetry use was not supported by an accepted set of clinical indications. The incidence of arrhythmias on nonindicated days was low (3.1 per 100 days of monitoring per nonindicated day),and the arrhythmias detected were clinically insignificant. Eliminating monitoring on nonindicated days could save a minimum of $53 per patient per day. The average 400-bed hospital with a conservative estimate of 5000 nonindicated patientdays per year could save $250,000 per year.

 Reducing the use of telemetry on nonindicated days may provide an opportunity for institutions to safely reduce cost as well as staff time and effort, while maintaining and potentially increasing patient safety.

We reviewed utilization patterns of inpatient cardiac telemetry at 4 academic medical centers. Our study highlights the potential overuse of inpatient cardiac telemetry and suggests that reducing inappropriate telemetry may improve the cost of care while maintaining patient safety.

		More than one-third of all telemetry days did not meet accepted indications for telemetry.

		The incidence of significant arrhythmia on nonindicated telemetry days was extremely small.

		Reducing the use of cardiac telemetry to only accepted indications may safely reduce healthcare costs.

With healthcare reform driving a heightened imperative for organizations to improve efficiency and reduce waste, it is timely and valuable to examine high-volume procedures and practices in order to identify overuse. At the same time, organizations must ensure that improved efficiency does not inadvertently reduce patient safety.

Cardiac disease or suspected cardiac disease is one of the most common reasons for which adults are admitted to the  hospital.

Many of these patients are placed on telemetry monitoring based on local practice or commonly accepted recommendations.

 Despite the well-defined indications for monitoring, developed by the American Heart Association in 2004,5 telemetry is often applied outside of these guidelines.

 A low incidence of clinically significant arrhythmias on  nonindicated days has been documented by Curry and colleagues as well as by others.

There are also well-documented cases of patient harm, including death, associated with alarm fatigue from excessive monitoring and false alarms.

 Telemetry is also expensive and labor intensive for hospitals. Equipment purchase, maintenance, and supplies, as well as nursing time,  are responsible for this expense. Finally, patient admissions may be delayed because of limited telemetry bed availability in many hospitals. The appropriate utilization of telemetry may decrease costs, reduce delays in admissions, and improve patient safety.

We undertook a multicenter study of the use of cardiac telemetry at 4 Massachusetts teaching hospitals to determine the utilization patterns, the frequency of telemetry use unsupported by guideline indications, the incidence of arrhythmias on both indicated and nonindicated days, and the estimated financial cost of nonindicated telemetry.

This study originated as a quality assurance initiative. Subsequently, the data collected for the quality assurance study were de-identified and became the basis for this research study. The institutional review board at 1 site acknowledged that this study was exempt because it involved only an analysis of retrospective, completely de-identified data.

Four hospitals representing urban academic medical centers in Massachusetts participated in the study. 

 shows the proportion of non—intensive care unit medical and surgical beds with telemetry capacity at each institution.

We reviewed the American College of Cardiology and the American Hospital Association guidelines

 for telemetry use and the local guidelines from the 4 participating institutions. We made minor modifications to the American College of Cardiology and American Hospital Association guidelines by classifying conditions into clear indications for telemetry monitoring and clear indications for discontinuing monitoring (

). We used the modified guidelines to designate each day of monitoring as indicated (meeting 1 or more guideline criteria) or nonindicated (not meeting any indication), based on criteria for initiating and discontinuing monitoring. We then designed an audit tool to support structured data abstraction from medical records. Data to be collected included start and stop date for monitoring episodes, guideline indication(s) for initiation of monitoring (if any), other reasons for instituting monitoring, number of days in each episode that met guideline criteria (indicated days), number of nonindicated days, and the occurrence of any significant arrhythmias during a monitoring episode, based on notes in the medical record. An arrhythmia was considered significant if it met 1 or more of these criteria: ventricular tachycardia of at least 3 continuous beats, supraventricular tachycardia of at least 3 beats including atrial fibrillation, any atrioventricular nodal block, and bradycardia with a heart rate of less than 50 beats per minute for more than 30 seconds. When the significance of an arrhythmia was not clear to an auditor, the auditor presented the finding to one of the authors, who made a final decision on significance.

Study subjects were all patients who were receiving care on general medical-surgical wards from January 1 through January 7, 2008, at the 4 participating health centers and who had at least 1 day of cardiac telemetry monitoring. All monitoring days were included in the abstraction and analysis, including days that occurred before January 1 and after January 7 when monitoring episodes began before or ended after the period from January 1 to January 7. Data abstraction was completed during the summer and fall of 2008. 

Estimates of the amount of nursing time required to manage cardiac monitors and costs of supplies (ie, monitor leads, batteries, paper for monitor printers) were obtained from nurse managers at each institution, who were asked to make estimates based on their own organization and their observation of nurses’ experience with the typical tasks related to the nurse management of cardiac monitoring equipment including placing, replacing, and repositioningmonitoring leads; orienting  patients to the monitor; responding to monitor alarms; running routine rhythm strips; and charting on monitor output after each shift. The average hourly pay for registered nurses was obtained from the 2010 Massachusetts Hospital Association’s 

 We added 20% to the estimated average base salary to cover the estimated average cost of benefits.To calculate nursing full-time equivalents (FTEs), we assumed an FTE nurse would be paid 40 hours per week for 52 weeks in a year (2080 hours per year).

To estimate the number of nonindicated days of monitoring in a typical year, we selected cases that began monitoring only during the study week (incident cases) and calculated the total number of nonindicated days during each related monitoring episode only in that week. We assumed that the number of incident cases and the duration of indicated and nonindicated monitoring in these cases were representative of all weeks in a typical year. We then multiplied the number of nonindicated days by 52 weeks to obtain an estimate of total annual nonindicated monitoring days. Adjusted cardiac arrhythmia event rates for indicated and nonindicated days and the significance of the difference in rates were obtained by fitting a generalized linear mixed model with a binomial distribution and logit link function, and using subjects and sites as random effects

 (a form of logistic regression). The odds ratio estimate of relative risk was obtained by exponentiating the fitted model parameter for the variable identifying whether days were indicated or not.

A 95% confidence interval (CI) for the odds ratio was obtained by taking the fitted model parameter for indication and adding and subtracting 1.96 standard errors for the parameter, and then exponentiating these end points. Analysis was performed using the Proc GLIMIX procedure

 with SAS statistical software, version 9.2 (SAS Institute Inc, Cary, North Carolina).

values associated with an effect or difference that were less than or equal to .05 indicated statistical significance. To compare medians, we used the Mann-Whitney U test. To compare proportions across sites, we used a χ2 test and adjusted multiple comparisons when indicated using a Bonferroni correction to the 

We identified 501 patients receiving care during the study period who had cardiac monitoring at any time during their admission at the 4 study sites: 42 at site 1, 131 at site 2, 145 at site 3, and 183 at site 4. Among these patients, there were 302 monitoring episodes (60.3%) that included only days on which there was always an acceptable indication for monitoring; 77 (15.4%) included only days on which there was no acceptable indication. The remaining 122 episodes (24.4%) typically started with 1 or more indicated days and ended with 1 or more nonindicated days. At site 1, there were no episodes consisting entirely of nonindicated days. Overall, the total number of patient days of monitoring was 1559, including 1011 (65%) indicated days and 548 (35%) nonindicated days. Of the 548 nonindicated days, 325 (59.3%) occurred during episodes with a mix of indicated and nonindicated days and 223.5 (40.6%) occurred during episodes comprising only nonindicated days.

The median duration of all episodes across all sites was 2 days (range, 0-18 days). There was modest but significant (

 <.0001) variation across sites, with median episode durationr anging from 1 day at site 4 to 4 days at site 1. A total of 430 (85.8%) episodes lasted 5 or fewer days, and only 17 (3.4%) were longer than 10 days.

Of the 501 monitoring episodes, 407 included 1 or more indicated days of monitoring. 

 shows the frequencies of telemetry indications. The most common indications for monitoring were suspected myocardial infarction (48.9% of episodes); arrhythmia, including atrial fibrillation (22.4%); and postcardiac procedures (16.2%). A total of 297 (73.0%) of the episodes with an indication were associated with only 1 indication for all or part of their monitoring episode, and 110 (27.0%) had more than 1 indication. For 326 episodes with an indication, information was available on adherence to guidelines for discontinuing monitoring. Telemetry was discontinued according to guidelines in 223 of those episodes (68.4%).

 shows the incidence rates of arrhythmias occurring on indicated and nonindicated days. Overall, the rate of arrhythmia was 20.7 per 100 indicated patient days and 3.1 per 100 nonindicated patient days. The unadjusted rate ratio comparing arrhythmia incidence on indicated and nonindicated days was 5.8 (95% CI, 3.58-9.40). The model-based, adjusted odds ratio comparing the arrhythmia rate on indicated days with that on nonindicated days was 4.48 (95% CI, 2.57- 8.17). No arrhythmias occurred during episodes comprising nonindicated days only. Most (80.2%) arrhythmias occurred during episodes comprising indicated days only. A total of 42 patients (8.4% of episodes) had more than 1 occurrence of arrhythmia, including 1 patient with 13 occurrences. Patients with more than 1 incident arrhythmia (n = 42) accounted for 70.5% of all arrhythmias.

At site 4, the 13 episodes of arrhythmia detected among 11 patients on nonindicated days included 6 episodes of tachycardia (heart rate, 121-151 beats per minute) and 4 episodes of bradycardia (heart rate, 43-48 beats per minute). In addition, there was 1 prolonged run of ventricular tachycardia, and 1 short (5 beats) run of ventricular tachycardia. At site 2, the single episode of arrhythmia on a nonindicated day was atrial fibrillation. At site 3, the 2 arrhythmias on nonindicated days were sinus tachycardia; at site 1, the 1 case was a single episode of bradycardia (heart rate, 45 beats per minute)

The average nurse manager estimate of total nursing time per patient each shift to manage a monitored patient was 30 minutes and the minimum was 15 minutes. Thus, the estimated minimum total daily nursing time over 3 shifts would be 45 minutes and the average would be 90 minutes per day for each monitored patient. The average hourly wage plus costs of benefits for nurses was $48.29 per hour. Therefore, the estimated minimum daily nursing cost was $36.22 (0.75 hour X $48.29 per hour) and the estimated average cost was $72.44 (1.5 hours X $48.29 per hour). The estimated minimum cost of monitoring supplies was $5 per day, and the average was $10 per day. Based on this information, we estimated the minimum total cost of 1 day of monitoring to be $53.29 and the average cost to be $82.44. To estimate total annual nonindicated hours of monitoring, we included onlypatients who had monitoring during the study week (January 1-7). For these cases, we then determined the number of nonindicated monitoring hours for each entire episode of monitoring. The number of these cases and related nonindicated days, respectively, were 14 and 6 (site 1), 76 and 25 (site 2), 68 and 33 (site 3), and 130 and 150 (site 4). We multiplied the number of nonindicated days in the study week by 52 to estimate total annual number of nonindicated days. The estimated annual number of nonindicated patient days per year varied for each site (312 for site 1, 1300 for site 2, 1716 for site 3, and 7800 for site 4). If each site eliminated nonindicated days and only had monitoring that met guidelines, the minimum amount saved per site ($16,626 for site 1; $69,277 for site 2; $91,446 for site 3; $415,662 for site 4) and the estimated minimum nursing FTEs saved (0.12 for site 1; 0.70 for site 2; 0.93 for site 3; 4.22 for site 4) were significant. Estimates of the average costs saving rather than the minimums would yield higher savings. The number of nonindicated days per monitored bed varied widely across the 4 institutions.

Cardiac telemetry has expanded in the past 30 years from its initial use for cardiac patients in the intensive care unit to include monitoring for lower-risk cardiac patients and even for noncardiac patients. Although there are well-established guidelines for telemetry use,

 our study and others have shown that many practitioners do not adhere to these recommendations. As with many frequently used technologies, this variability has both positive and negative effects. The literature is replete with economic analyses demonstrating variations in healthcare costs across the country, as well as a lack of correlation between these expenses and improved quality.

This retrospective study at 4 teaching hospitals revealed that cardiac telemetry was being utilized in many patients without a guideline-supported indication. We found that 35% of all cardiac telemetry days were considered nonindicated days, with a range from 20% to 40% of all days at across study sites. We identified 2 types of overuse: nonindicated initiation of telemetry and continuation past the guidelinerecommended stop date. The economic impact of use of telemetry without a guideline-based indication was significant. Based on data for nursing time and supplies alone, we estimated that the average annualized excess cost for a 400-bed hospital with 5000 nonindicated days per year based on a 35% rate of nonindicated cardiac telemetry was $250,000 per year. In addition, for many hospitals, access to telemetry beds impacts the ability to admit patients from the emergency department, resulting in flow bottlenecks and backups. The opportunity cost resulting from improved patient flow would add to these savings.

Another major aspect of the overuse of technology relates to the question of improved safety versus reduced safety. Cardiac telemetry use represents such an issue. Our study demonstrated that there were a significant number of arrhythmic events in patients who met the criteria for telemetry and many fewer in patients who did not. This circumstance has been demonstrated in a number of prior publications. Kanwar and colleagues studied patients at a single hospital and determined that on a per patient level, more than 50% of patients received cardiac telemetry that was not indicated.

 Comparing the rates of arrhythmic events on indicated and nonindicated telemetry days, Curry and colleagues determined that when the guidelines derived from the American College of Cardiology policy statement were followed for those patients with nonindicated days, the rate of  arrhythnmias was very low at 2.6 per 100 monitored days, which is very close to our low rate of arrhythmias on nonindicated days.7 Our study also documented the safety of closely adhering to agreed-on guidelines for cardiac telemetry. We determined that the rate of arrhythmia on nonindicated days was low—about one-sixth the rate on indicated days. While the rate of arrhythmias was not zero for nonindicated days, the severity of the arrhythmias on nonindicated days was quite low and more often asymptomatic.

One could argue that missing any opportunity to detect a serious cardiac arrhythmia is a concern. Although our study did not have a control arm to determine the incidence of significant arrhythmias in a “healthy” inpatient population, there is indirect evidence from the cardiology literature on Holter monitoring that demonstrates that arrhythmias can be identified with monitoring even in apparently healthy populations.

 So it is possible that the rate of arrhythmias occurring on nonindicated days in monitored patients may be comparable to the rate in hospitalized patients who have not been monitored. A formal costeffectiveness analysis is beyond the scope of this study, but given the cost of monitoring that we documented and the questionable clinical significance of the arrhythmias documented, it is likely that monitoring without an indication would not be cost-effective compared with other evidencebased hospital service.

From a safety perspective, the overuse of telemetry has significant implications. First, there is good evidence that alert fatigue and interruptions negatively impact quality.

 The workgenerated by monitoring nonindicated patients creates both false alarms and interruptions for nurses. Second, this “noise” distracts staff from paying closer attention to the indicated patients, and there are cases in the literature and the press where adverse outcomes have resulted. Last, one could argue that the time required to attend to monitoring nonindicated patients takes away from nurses’ availability for both other patient needs and face time, which also leads to better quality.

We found wide variation in the number of nonindicated days per monitored bed across sites and a modest amount of variation in the duration of episodes. The variation may be the result of local clinician practices, availability of monitored beds (which differs by organization), case mix, or other factors that we did not address in this study. Site 1 had fewer telemetry beds than the other sites and also had the lowest number of nonindicated days per bed. This suggests that it may be possible to improve adherence to guidelines by limiting the availability of telemetry, although the association between limited telemetry availability and guideline adherence is not definitive evidence that limiting telemetry would result in improved guideline adherence.

There are significant limitations to our study. We studied 4 urban hospitals for only 1 week retrospectively and reviewed a relatively small number of patients. This review may have underestimated or overestimated the true rate of indicated versus nonindicated use of cardiac telemetry, as well as the estimates of annual nonindicated days and related costs, especially if there was significant seasonal or month-to-month variation in telemetry use. In addition, even though we saw a statistically significant difference in the rate of arrhythmias on indicated versus nonindicated days, we did not determine the effect on actual outcomes (ie, mortality) of these patients. It is likely, however, that given the significantly higher rate and severity of arrhythmias in the indicated group, this observation would hold in a larger study. Although all 4 hospitals were urban hospitals in Massachusetts, the types of cases are overwhelmingly primary, not secondary or tertiary care. Among the cases we studied, 12% involved preprocedure or postprocedure patients and 88% involved general cardiac and medical patients. We studied patients during only 1 week of the year and then extrapolated the rates of nonindicated days across an entire year. As a result, we may have overestimated or underestimated the nonindicated days, yet the cases we included represented a broad array of  common cardiac diagnoses that have little seasonal variation.

Our economic analysis was limited in scope. There are indirect  economic impacts of the volume of telemetry on equipment, maintenance of equipment, and other costs that we did not include. In addition, we did not take into account the economic impact of impaired patient flow resulting from access to telemetry beds. Our cost-savings estimate assumed that nursing staffing could be adjusted to accommodate the reduced demand. That being said, even with our minimalist economic projections, the cost of monitoring nonindicated patients is substantial, both to an individual hospital and in aggregate for the nation.

A strength of the study is that data from 4 separate hospitals with different individual patterns of utilization yielded similar trends. We looked at all patients outside of the intensive care units at each institution, and so did not exclude any subjects. The literature contains evidence of the lack of value of monitoring patients who do not meet the American Hospital Association criteria, and our study supports that conclusion. In addition, and perhaps more importantly, our study shines a light on the financial and patient care implications of over-utilization of telemetry monitoring.

Others have studied the effectiveness of implementing quality improvement programs to reduce unnecessary cardiac telemetry successfully.

 The initiation of quality improvement and more intensely managed telemetry programs have the potential to reduce the nonindicated utilization of telemetry monitoring. Computerized order entry with decision tools and clinician education could have a significant impact on the correct utilization of telemetry. More research to implement better systems to promote adherence to guidelines is necessary.

Cardiac telemetry monitoring may be overused in US hospitals and can be safely reduced, resulting in cost savings to healthcare system. Efforts to reduce utilization need to be explored.

 From Baystate Health, Inc (EMB), Springfield, MA; UMass Memorial Health Care (RAK, RL), Worcester, MA; Tufts Medical Center (DGF), Boston, MA; Mount Auburn Hospital (SAA), Cambridge, MA

 The authors (EMB, RAK, RL, DGF, SAA) report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

 Concept and design (EMB, RAK, DGF, SAA);acquisition of data (EMB, RAK, DGF, SAA); analysis and interpretation of data (EMB, RAK, RL, DGF); drafting of the manuscript (EMB, RAK, RL, DGF, SAA); critical revision of the manuscript for important intellectual content (EMB, RAK, RL, DGF, SAA); statistical analysis (EMB, RAK, RL); provision of study materials or patients (EMB, SAA); and administrative, technical, or logistic support (EMB, RAK).

 Evan M. Benjamin, MD, Baystate Medical Center, 759 Chestnut St, SG618, Springfield, MA 01199. E-mail: Evan.Benjamin@bhs.org.1. Boie ET. Initial evaluation of chest pain. Emerg Med Clin North Am. 2005;23(4):937-957.

2. Pitts SR, Niska RW, Xu J, Burt CW. National Hospital Ambulatory Medical Care Survey: 2006 emergency department summary. Natl Health Stat Report. 2008;(7):1-38.

3. Hollander JE, Sites FD, Pollack CV Jr, Shofer FS. Lack of utility of telemetry monitoring for identification of cardiac death and  life-threatening ventricular dysrhythmias in low-risk patients with chest pain. Ann Emerg Med. 2004;43(1):71-76.

4. Braunwald E, Antman EM, Beasley JW, et al; American College of Cardiology; American Heart Association. Committee on the Management of Patients With Unstable Angina. ACC/AHA 2002 guideline update for the management of patients with unstable angina and non- ST-segment elevation myocardial infarction—summary article: a report of the American College of Cardiology/American Heart Association task force on practice guidelines (Committee on the Management of Patients With Unstable Angina). J Am Coll Cardiol. 2002 2;40(7):1366-1374.

Reducing inappropriate use of cardiac telemetry may improve the cost of care while maintaining patient safety.

Impact of Cardiac Telemetry on Patient Safety and Cost

To identify insights gained from Hawaii’s experience with healthcare costs in an environment of mandated employer-based coverage and implications for other states as implementation of the Affordable Care Act ramps up.

We reviewed literature on healthcare costs in Hawaii and analyzed descriptive statelevel and national data from a variety of sources, including MEPS and Hawaii’s Department of Commerce and Consumer Affairs. In addition, we conducted 14 interviews with 12 stakeholder organizations in Hawaii’s healthcare market, including representatives of health plans, delivery systems, physicians, employers, government, the non-profit sector, and academia. 

After almost 40 years of mandated employer-based coverage, Hawaii has a lower uninsured rate than other states and lower employer-based premiums. Stakeholders we interviewed attributed lower costs in part to the concentrated insurance market, in which the dominant carrier acts largely like a single payer in the commercial market; stakeholders raised administrative cost efficiencies and negotiating clout as factors contributing to lower employerbased premiums. While premiums are lower than the US average, the rate of growth in costs is not. As a result, Hawaii’s healthcare stakeholders are focused on aligning incentives for payers and providers.

Based on interviews with stakeholders and the review of current data, we conclude that mandated coverage will not slow the inexorable rise in healthcare costs or solve the growing affordability issues for health insurance in the United States. To expand access to coverage in a manner that is sustainable over time, healthcare cost growth must be addressed.

 The State of Hawaii has mandated employer-based coverage since 1974, offering an opportunity for learning as implementation of the Affordable Care Act ramps up.

		Hawaii’s uninsured rate is substantially lower than the US average, with the differential almost entirely associated with employment-based coverage.

		Hawaii features relatively low employer-based premiums, explained in part by a concentrated insurance market that yields lower administrative costs.

		The mandate has had little effect on the growth rate in the cost of coverage, which mirrors the rest of the United States. Hawaii’s stakeholders are focused on delivery and reimbursement approaches that maintain quality while reducing costs.

As states across the nation gear up to implement provisions of the Affordable Care Act (ACA), there is an opportunity to reflect on lessons learned from the state of Hawaii’s nearly 40 years of experience with an employer mandate. Hawaii’s mandate has indeed expanded coverage, and the state continues to have a relatively low proportion of uninsured residents, even as that figure has increased in past years. Furthermore, Hawaii’s employer-based health insurance premiums are among the lowest in the country—in spite of requiring coverage of a relatively generous set of benefits with limited employee cost sharing. At the same time, the rate of cost growth observed in the state is no lower than the US average&mdash;meaning that Hawaii’s residents face the same sustainability issues faced by states without a mandate and signifying that payment and delivery system reform are critical if we are to have a sustainable healthcare system. To gain insights from Hawaii’s experience, we reviewed the literature on healthcare costs in Hawaii, analyzed descriptive state level and national data, and conducted a series of interviews with key stakeholders in Hawaii’s healthcare system. Interviewees included representatives of the Governor's Office, Hawaii Department of Commerce and Consumer Affairs, Hawaii branch for the National Federation of Independent Businesses, Hawaii Health Connector, Hawaii Health Information Exchange, Hawaii Independent Physicians Association, Hawaii Medical Service Association (HMSA), Hawaii Pacific Health, Hawaii Prepaid Health Care Advisory Council, Hawaii State Senate, Kaiser Foundation Health Plan, The Queen’s Medical Center, and University of Hawaii School of Nursing. Interviews lasted 45 to 90 minutes and took place between January and September 2012; 12 were conducted in person and 2 by phone.  

Hawaii’s Employer Mandate: Overview and Effects

The state of Hawaii has mandated employer-based coverage since 1974, offering an opportunity for learning as implementation of the Affordable Care Act ramps up.

June 2013

Consumer Cost Sharing and Use of Biopharmaceuticals for Rheumatoid Arthritis

Effects of Medicare Part D Coverage Gap on Medication Adherence

Impact of Cardiac Telemetry on Patient Safety and Cost

Affordability in a Mandated Environment

Medicare and Commercial Inpatient Resource Use: Impact of Hospital Competition

Shifting the Dialogue From Hospital Readmissions to Unplanned Care

Population Health Approach for Diabetic Patients With Poor A1C Control

Cervical Cancer Screening Overuse and Underuse: Patient and Physician Factors

Estimating the Staffing Infrastructure for a Patient-Centered Medical Home

Identifying Groups of Nonparticipants in Type 2 Diabetes Mellitus Education

Accountable Care Organizations: How to Dress for Success