Low-Cost Generic Program Use by Medicare Beneficiaries: Implications for Medication Exposure Misclassification in Administrative Claims Data

The quality of mental health care delivered to patients with schizophrenia and related disorders in the Italian mental health system. The QUADIM project: a multi-regional Italian investigation based on healthcare utilisation databases

AIMS

To evaluate the quality of mental health care delivered to patients with schizophrenia and related disorders taken-in-care by mental health services in four Italian regions (Lombardy, Emilia-Romagna, Lazio, Sicily).

METHODS

Thirty-one clinical indicators concerning accessibility, appropriateness, continuity and safety were defined and estimated using healthcare utilisation (HCU) databases, containing data on mental health treatments, hospital admissions, outpatient interventions, lab tests and drug prescriptions.

RESULTS

A total of 70 586 prevalent patients with schizophrenia and related disorders treated in 2015 were identified, of whom 1752 were newly taken-in-care by the facilities of regional mental health services. For most patients community care was accessible and moderately intensive. However, care pathways were not implemented based on a structured assessment and only half of the patients received psychosocial treatments. One patient out of ten had access to psychological interventions and psychoeducation. Activities specifically addressed to families involved a third of prevalent patients and less than half of new patients. One patient out of six was admitted to a community residential facility, and one out of ten to a General Hospital Psychiatric Ward (GHPW); higher values were identified in new cases. In general hospitals, few patients had a length of stay (LoS) of more than 30 days, while one-fifth of the admissions were followed by readmission within 30 days of discharge. For two-thirds of patients, continuity of community care was met, and six times out of ten a discharge from a GHPW was followed by an outpatient contact within 2 weeks. For cases newly taken-in-care, the continuity of community care was uncommon, while the readiness of outpatient contacts after discharge was slightly more frequent. Most of the patients received antipsychotic medication, but their adherence to long-term treatment was low. Antipsychotic polytherapy was frequent and the control of metabolic side effects was poor. The variability between regions was high and consistent in all the quality domains.

CONCLUSIONS

The Italian mental health system could be improved by increasing the accessibility to psychosocial interventions, improving the quality of care for newly taken-in-care patients, focusing on somatic health and mortality, and reducing regional variability. Clinical indicators demonstrate the strengths and weaknesses of the mental health system in these regions, and, as HCU databases, they could be useful tools in the routine assessment of mental healthcare quality at regional and national levels.

Does the mental health system provide effective coverage to people with schizophrenic disorder? A self-controlled case series study in Italy

PURPOSE

To measure indicators of timeliness and continuity of treatments on patients with schizophrenic disorder in 'real-life' practice, and to validate them through their relationship with relapse occurrences.

METHODS

The target population was from four Italian regions overall covering 22 million beneficiaries of the NHS (37% of the entire Italian population). The cohort included 12,054 patients newly taken into care for schizophrenic disorder between January 2015 and June 2016. The self-controlled case series (SCCS) design was used to estimate the incidence rate ratio of relapse occurrences according to mental healthcare coverage.

RESULTS

Poor timeliness (82% and 33% of cohort members had not yet started treatment with psychosocial interventions and antipsychotic drug therapy within the first year after they were taken into care) and continuity (27% and 23% of patients were persistent with psychosocial interventions, and antipsychotic drug therapy within the first 2 years after starting the specific treatment) were observed. According to SCCS design, 4794 relapses occurred during 9430 PY (with incidence rate of 50.8 every 100 PY). Compared with periods not covered by mental healthcare, those covered by psychosocial intervention alone, antipsychotic drugs alone and by psychosocial intervention and antipsychotic drugs together were, respectively, associated with relapse rate reductions of 28% (95% CI 4-46%), 24% (17-30%) and 44% (32-53%).

CONCLUSION

Healthcare administrative data may contribute to monitor and to assess the effectiveness of a mental health system. Persistent use of both psychosocial intervention and antipsychotic drugs reduces risk of severe relapse.

Cardiovascular Outcomes and Mortality Associated With Discontinuing Statins in Older Patients Receiving Polypharmacy

IMPORTANCE

Polypharmacy is a major health concern among older adults. While deprescribing may reduce inappropriate medicine use, its effect on clinical end points remains uncertain.

OBJECTIVE

To assess the clinical implications of discontinuing the use of statins while maintaining other drugs in a cohort of older patients receiving polypharmacy.

DESIGN, SETTING, AND PARTICIPANTS

This retrospective, population-based cohort study included the 29 047 residents in the Italian Lombardy region aged 65 years or older who were receiving uninterrupted treatment with statins, blood pressure-lowering, antidiabetic, and antiplatelet agents from October 1, 2013, until January 31, 2015, with follow-up through June 30, 2018. Data were collected using the health care utilization database of Lombardy region in Italy. Data analysis was conducted from March to November 2020.

EXPOSURES

Cohort members were followed up to identify those who discontinued statins. Among this group, those who maintained other therapies during the first 6 months after statin discontinuation were 1:1 propensity score matched with patients who discontinued neither statins nor other drugs.

MAIN OUTCOME AND MEASURES

The pairs of patients discontinuing and maintaining statins were followed up from the initial discontinuation until June 30, 2018, to estimate the hazard ratios (HRs) and 95% CIs for fatal and nonfatal outcomes associated with statin discontinuation.

RESULTS

The full cohort inclued 29 047 patients exposed to polypharmacy (mean [SD] age, 76.5 [6.5] years; 18 257 [62.9%] men). Of them, 5819 (20.0%) discontinued statins while maintaining other medications, and 4010 (68.9%) of them were matched with a comparator. In the discontinuing group, the mean (SD) age was 76.5 (6.4) years, 2405 (60.0%) were men, and 506 (12.6%) had Multisource Comorbidity Scores of 4 or 5. In the maintaining group, the mean (SD) age was 76.1 (6.3) years, 2474 (61.7%) were men, and 482 (12.0%) had multisource comorbidity scores of 4 or 5. Compared with the maintaining group, patients in the discontinuing group had increased risk of hospital admissions for heart failure (HR, 1.24; 95% CI, 1.07-1.43) and any cardiovascular outcome (HR, 1.14; 95% CI, 1.03-1.26), deaths from any cause (HR, 1.15; 95% CI, 1.02-1.30), and emergency admissions for any cause (HR, 1.12; 95% CI, 1.05-1.19).

CONCLUSIONS AND RELEVANCE

In this study of patients receiving polypharmacy, discontinuing statins while maintaining other drug therapies was associated with an increase in the long-term risk of fatal and nonfatal cardiovascular outcomes.

Association of Oral Anticoagulants and Verapamil or Diltiazem With Adverse Bleeding Events in Patients With Nonvalvular Atrial Fibrillation and Normal Kidney Function

IMPORTANCE

Direct oral anticoagulants (DOACs) are purported to have fewer drug-drug interactions than warfarin. However, potential interactions with coprescribed medications are still a safety concern. Verapamil hydrochloride and diltiazem hydrochloride are combined P-glycoprotein (P-gp) and CYP3A4 inhibitors and may be associated with increases in the risk of bleeding with DOACs.

OBJECTIVE

To evaluate the risk of bleeding with DOACs and verapamil or diltiazem using an active comparator design.

DESIGN, SETTING, AND PARTICIPANTS

A comparative effectiveness active comparator cohort study was conducted using US population-based data (2010-2015) analyzed between January 1 and July 15, 2019. Data were obtained on 48 442 patients with nonvalvular atrial fibrillation who had received an index prescription of dabigatran, rivaroxaban, or apixaban between October 19, 2010, through June 30, 2015, with final follow-up on October 1, 2015. Analysis was restricted to individuals with no history of kidney disease who were receiving standard doses of the DOACs.

EXPOSURES

Patients with initial prescriptions of DOACs who were receiving verapamil or diltiazem were compared with those receiving amlodipine or metoprolol.

MAIN OUTCOMES AND MEASURES

Overall and gastrointestinal major, moderate, and minor bleeding using primary or secondary diagnoses. Hazard ratios and 95% CIs were estimated using inverse probability of treatment weights in Cox proportional hazards regression models.

RESULTS

Of the 48 442 patients reviewed, analysis was conducted on 1764 patients receiving DOACs with verapamil or diltiazem compared with 3105 receiving amlodipine and 1793 patients receiving DOACs with verapamil or diltiazem compared with 3224 receiving metoprolol. Depending on the comparison, approximately 60% of the cohort were younger than 65 years and male, which differed by treatment group. Rivaroxaban and apixaban were not associated with increased rates of bleeding for patients receiving verapamil or diltiazem compared with those receiving amlodipine or metoprolol. Among patients receiving dabigatran etexilate, the overall bleeding rate was 52% higher (hazard ratio, 1.52; 95% CI, 1.05-2.20) with verapamil or diltiazem vs amlodipine and 43% higher (hazard ratio, 1.43; 95% CI, 1.02-2.00) vs metoprolol. Bleeding rates for dabigatran with verapamil or diltiazem were higher overall for other bleeding types (244.9 vs 158.4 per 1000 person-years; adjusted hazard ratios of overall GI bleeding: 2.16; 95% CI, 1.30-3.60; minor bleeding: 1.56; 95% CI, 1.07-2.27; and minor GI bleeding: 2.16; 95% CI, 1.29-3.63). Sensitivity analyses showed consistent results for dabigatran when used with verapamil and diltiazem, with magnitudes ranging from 50% to 100% increased hazard rates and no significant results for apixaban or rivaroxaban.

CONCLUSIONS AND RELEVANCE

Current US prescribing information only recommends prescribing changes with dabigatran and P-gp inhibitors with lower kidney function. This study found increased bleeding risk associated with dabigatran when used concomitantly with the P-gp inhibitors verapamil and diltiazem in individuals with normal kidney function. Clinicians and patients may need to consider these drug-drug interactions when choosing oral anticoagulation.

Real-world adherence for direct oral anticoagulants in a newly diagnosed atrial fibrillation cohort: does the dosing interval matter?

Background

Differences in adherence may represent drug properties (e.g. dosing interval) or patient experiences while on treatment. Adherence to direct oral anticoagulants (DOACs) in nonvalvular atrial fibrillation (NVAF) is important to maintain effectiveness over the course of treatment.

Methods

This was a retrospective cohort study using 2009–2015 Truven Health MarketScan Databases. New initiators of dabigatran, rivaroxaban, and apixaban with NVAF were identified. Twelve months of continuous enrollment before treatment was required to assess demographics and medical history. Proportion of days cover (PDC) was used to measure adherence at 3, 6, 9 and 12-month. Gaps in therapy and treatment switches were also evaluated. Logistic regression was used to compare high adherence (PDC ≥0.80).

Results

A total of 14,864 dabigatran, 16,005 rivaroxaban, and 8078 apixaban users were identified. Apixaban users had the highest adherence overall, with mean PDC at 3, 6, 9, and 12-months of 0.83, 0.76, 0.72, and 0.69, while dabigatran had the lowest adherence of 0.78, 0.67, 0.61, and 0.57. Adherence to DOACs increased with increased stroke risk scores. Adherence was also higher when first days supplied was > 30 days compared to 30 days and when filled via mail order pharmacies. Switching was highest among dabigatran users. Apixaban users were the most likely to have high adherence versus dabigatran (OR = 1.73, 95% CI = 1.60–1.88) and versus rivaroxaban (OR = 1.24, 95% CI = 1.14–1.34) at 12-months.

Conclusions

Apixaban users had the highest overall adherence despite twice-daily dosing versus once-daily dosing for rivaroxaban. These findings can be useful for formulary decision-making and when assessing treatment options.

Bias in pharmacoepidemiologic studies using secondary health care databases: a scoping review

BACKGROUND

The availability of clinical and therapeutic data drawn from medical records and administrative databases has entailed new opportunities for clinical and epidemiologic research. However, these databases present inherent limitations which may render them prone to new biases. We aimed to conduct a structured review of biases specific to observational clinical studies based on secondary databases, and to propose strategies for the mitigation of those biases.

METHODS

Scoping review of the scientific literature published during the period 2000-2018 through an automated search of MEDLINE, EMBASE and Web of Science, supplemented with manually cross-checking of reference lists. We included opinion essays, methodological reviews, analyses or simulation studies, as well as letters to the editor or retractions, the principal objective of which was to highlight the existence of some type of bias in pharmacoepidemiologic studies using secondary databases.

RESULTS

A total of 117 articles were included. An increasing trend in the number of publications concerning the potential limitations of secondary databases was observed over time and across medical research disciplines. Confounding was the most reported category of bias (63.2% of articles), followed by selection and measurement biases (47.0% and 46.2% respectively). Confounding by indication (32.5%), unmeasured/residual confounding (28.2%), outcome misclassification (28.2%) and "immortal time" bias (25.6%) were the subcategories most frequently mentioned.

CONCLUSIONS

Suboptimal use of secondary databases in pharmacoepidemiologic studies has introduced biases in the studies, which may have led to erroneous conclusions. Methods to mitigate biases are available and must be considered in the design, analysis and interpretation phases of studies using these data sources.

A systematic review of methods for determining cross-sectional active medications using pharmacy databases

PURPOSE

Pharmacy dispensing databases are often used to identify patients' medications at a particular time point, for example to measure prescribing quality or the impact of medication use on clinical outcomes. We performed a systematic review of studies that examined methods to assess medications in use at a specific point in time.

METHODS

Comprehensive literature search to identify studies that compared active medications identified using pharmacy databases to medications identified using nonautomated data sources. Two investigators independently reviewed abstracts and full-text material.

RESULTS

Of 496 studies screened, 29 studies evaluating 50 comparisons met inclusion criteria. Twenty-nine comparisons evaluated fixed look-back period approaches, defining active medications as those filled in a specified period prior to the index date (range 84-730 days). Fourteen comparisons evaluated medication-on-hand approaches, defining active medications as those for which the most recent fill provided sufficient supply to last through the study index date. Sensitivity ranged from 48% to 93% for fixed look-back period approaches and 35% to 97% for medication-on-hand approaches. Interpretation of comparative performance of methods was limited by use of different reference sources, target medication classes, and databases across studies. In four studies with head-to-head comparisons of these methods, sensitivity of the medication-on-hand approach was a median of 7% lower than the corresponding fixed look-back approach.

CONCLUSIONS

The reported accuracy of methods for identifying active medications using pharmacy databases differs greatly across studies. More direct comparisons of common approaches are needed to establish the accuracy of methods within and across populations, medication classes, and databases.

The relative burden of community-acquired pneumonia hospitalizations in older adults: a retrospective observational study in the United States

Background

The risk of community-acquired pneumonia (CAP) increases with age and significantly impacts morbidity and mortality in the elderly population. The burden of illness and cost of preventing CAP has not been compared to other serious diseases.

Methods

This retrospective analysis used claims data from 2014 to 2015 and compared hospitalizations for CAP, myocardial infarction (MI), stroke, and osteoporotic fractures (OF) in adults aged ≥65 years enrolled in a Medicare Advantage insurance plan. Individuals who had not already been hospitalized for one of these conditions and did not have evidence of long-term care were included in the study. Hospitalizations for each condition were described by length of stay, readmissions, mortality, and total costs. Preventive measures included vaccinations for CAP and medications for MI, stroke, and OF.

Results

A total of 1,949,352 individuals were included in the cohort. In 2015, the rate of CAP-related hospitalizations was the highest at 846.7 per 100,000 person-years compared to 405 for MI, 278.9 for stroke, and 343.9 for OF. Vaccination costs for CAP were $40.2 million including $14.1 million for pneumococcal and $26.1 million for influenza vaccines. The cost of preventive medications for MI and stroke reached over $661 million and OF totaled $169 million.

Conclusions

Although CAP has a higher burden of hospitalization and total costs than MI, stroke, and OF in the elderly population, prevention efforts were disproportionately smaller for CAP. Prioritization of CAP prevention is needed to substantially reduce the burden of CAP.

Adherence to Rivaroxaban, Dabigatran, and Apixaban for Stroke Prevention for Newly Diagnosed and Treatment-Naive Atrial Fibrillation Patients: An Update Using 2013-2014 Data

BACKGROUND

Few studies have assessed adherence to non-vitamin K antagonist oral anticoagulants (NOACs), especially using contemporary data now that multiple NOACs are available.

OBJECTIVE

To compare adherence and treatment patterns among NOACs for stroke prevention in patients with nonvalvular atrial fibrillation (NVAF).

METHODS

Incident and treatment-naive NVAF patients were identified during 2013-2014 from a large claims database in this retrospective cohort study. Patients were included who initiated rivaroxaban, dabigatran, or apixaban within 30 days after diagnosis. Adherence to the index medication and adherence to any oral anticoagulant was assessed using the proportion of days covered (PDC) at 3, 6, and 9 months. The number of switches and gaps in therapy were also evaluated. Analyses were stratified by stroke risk scores, and a logistic regression model was used to control for factors that may predict high adherence.

RESULTS

Dabigatran had lower adherence (PDC = 0.76, 0.64, 0.57) compared with rivaroxaban (PDC = 0.83, 0.73, 0.66; P < 0.001) and apixaban (PDC = 0.82, 0.72, 0.66; P < 0.001) at 3, 6, and 9 months of follow-up and twice the number of switches to either other anticoagulants or antiplatelet therapy. Adherence was higher overall as stroke risk increased, and dabigatran had consistently lower adherence compared with the other NOACs. Multivariable logistic regression predicting PDC ≥ 0.80 showed rivaroxaban users with higher odds of high adherence compared with dabigatran or rivaroxaban across all time periods. Adjusted analyses showed that increasing age and comorbid hypertension and diabetes were associated with higher adherence.

CONCLUSIONS

In this real-world analysis of adherence to NOACs, rivaroxaban and apixaban had favorable unadjusted adherence profiles compared with dabigatran, while rivaroxaban users had higher odds of high adherence (PDC ≥ 0.80) among the NOACs in adjusted analyses. Clinicians and managed care organizations should consider the implications of lower adherence on clinical outcomes and quality assessment.

DISCLOSURES

This project was supported by the National Center for Advancing Translational Sciences, National Institutes of Health, through grant number UL1TR000117. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. The authors have nothing to disclose. Study concept and design were contributed by Brown and Shewale. Brown and Talbert collected the data, and data analysis was performed primarily by Brown, along with Shewale and Talbert. The manuscript was written primarily by Brown, along with Shewale, and revised by all the authors.

Quality of Antiepileptic Treatment Among Older Medicare Beneficiaries With Epilepsy: A Retrospective Claims Data Analysis

BACKGROUND

Enzyme-inducing antiepileptic drugs (EI-AEDs) are not recommended for older adults with epilepsy. Quality Indicator for Epilepsy Treatment 9 (QUIET-9) states that new patients should not receive EI-AEDs as first line of treatment. In light of reported racial/ethnic disparities in epilepsy care, we investigated EI-AED use and QUIET-9 concordance across major racial/ethnic groups of Medicare beneficiaries.

RESEARCH DESIGN

Retrospective analyses of 2008-2010 Medicare claims for a 5% random sample of beneficiaries 67 years old and above in 2009 augmented for minority representation. Logistic regressions examined QUIET-9 concordance differences by race/ethnicity adjusting for individual, socioeconomic, and geography factors.

SUBJECTS

Epilepsy prevalent (≥1 International Classification of Disease-version 9 code 345.x or ≥2 International Classification of Disease-version 9 code 780.3x, ≥1 AED), and new (same as prevalent+no seizure/epilepsy events nor AEDs in 365 d before index event) cases.

MEASURES

Use of EI-AEDs and QUIET-9 concordance (no EI-AEDs for the first 2 AEDs).

RESULTS

Cases were 21% white, 58% African American, 12% Hispanic, 6% Asian, 2% American Indian/Alaskan Native. About 65% of prevalent, 43.6% of new cases, used EI-AEDs. QUIET-9 concordance was found for 71% Asian, 65% white, 61% Hispanic, 57% African American, 55% American Indian/Alaskan new cases: racial/ethnic differences were not significant in adjusted model. Beneficiaries without neurology care, in deductible drug benefit phase, or in high poverty areas were less likely to have QUIET-9 concordant care.

CONCLUSIONS

EI-AED use is high, and concordance with recommendations low, among all racial/ethnic groups of older adults with epilepsy. Potential socioeconomic disparities and drug coverage plans may affect treatment quality and opportunities to live well with epilepsy.

Adherence to Rivaroxaban, Dabigatran, and Apixaban for Stroke Prevention in Incident, Treatment-Naïve Nonvalvular Atrial Fibrillation

BACKGROUND

Non-vitamin K antagonist oral anticoagulants (NOACs) are widely used for prevention of stroke secondary to nonvalvular atrial fibrillation (NVAF). Increased use of NOACs is partially a result of simplified regimens compared with warfarin, which has been associated with poor adherence and persistence to therapy. Few studies have assessed adherence to NOACs, especially using contemporary data now that multiple NOACs are available.

OBJECTIVE

To evaluate adherence to NOACs in a cohort of newly diagnosed NVAF patients who are commercially insured.

METHODS

Incident, treatment-naïve NVAF patients were identified in 2013 from a large claims database. Patients were included who initiated rivaroxaban, dabigatran, or apixaban within 30 days after diagnosis. Subjects were required to have 12 months of pre-index information to assess demographic and clinical characteristics (comorbidities, CHA₂ DS₂-VASc, and HAS-BLED scores). Adherence to the index medication and adherence to any oral anticoagulant was assessed using proportion of days covered (PDC) at 3, 6, and 9 months. The number of switches and gaps in therapy were also evaluated. Analyses were stratified by stroke risk scores, and a logistic regression model was used to control for factors that may predict high adherence (PDC ≥ 0.80).

RESULTS

A total of 3,455 rivaroxaban, 1,264 dabigatran, and 504 apixaban users were included with no major clinical or demographic differences between groups. At 3, 6, and 9 months of follow-up, dabigatran had lower adherence (PDC = 0.77, 0.67, and 0.62) compared with rivaroxaban (PDC = 0.84, 0.75, and 0.70; P < 0.001) and apixaban (PDC = 0.82, 0.75, and 0.71; P < 0.001), as well as nearly twice the number of switches to either other anticoagulants or antiplatelet therapy. At 9 months, 55.0% of rivaroxaban initiators had PDC ≥ 0.80, which was comparable with 56.8% for apixaban and significantly greater than 46.7% for dabigatran (P < 0.001). Adherence was higher overall as stroke risk increased and showed dabigatran had consistently lower adherence compared with the other NOACs. Overall adherence to any oral anticoagulants, allowing for switches to another NOAC or warfarin, was not dependent on the index medication (9-month PDC = 0.74, 0.71, and 0.74 for rivaroxaban, dabigatran, and apixaban initiators). Adjusted analyses showed that increasing age and comorbid hypertension and diabetes were associated with higher adherence. Compared with rivaroxaban, dabigatran initiators had nearly 30% lower odds of being adherent to their index medication, and no differences were observed between apixaban and rivaroxaban. At 9 months, there were no differences between NOACs for overall adherence to oral anticoagulants.

CONCLUSIONS

In this real-world analysis of adherence to NOACs, rivaroxaban and apixaban had favorable profiles compared with dabigatran, and rivaroxaban appeared to have higher overall adherence among the NOACs. Clinicians and managed care organizations should consider the implications of lower adherence on clinical outcomes as well as quality assessment.

DISCLOSURES

The project described was supported by the National Center for Advancing Translational Sciences, National Institutes of Health, through grant number UL1TR000117. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. Brown reports receiving a training fellowship from Human and Pfizer. Study concept and design were contributed by Brown and Shewale. Talbert took the lead in data collection, along with Brown, and data interpretation was primarily performed by Brown, along with Shewale. The manuscript was written primarily by Brown, along with Shewale, and revised by all the authors.

Empirical Assessment of the Impact of Low-Cost Generic Programs on Adherence-Based Quality Measures

In the United States, federally-funded health plans are mandated to measure the quality of care. Adherence-based medication quality metrics depend on completeness of administrative claims data for accurate measurement. Low-cost generic programs (LCGPs) cause medications fills to be missing from claims data as medications are not adjudicated through a patient’s insurance. This study sought to assess the magnitude of the impact of LCGPs on these quality measures. Data from the 2012–2013 Medical Expenditure Panel Survey (MEPS) were used. Medication fills for select medication classes were classified as LCGP fills and individuals were classified as never, sometimes, and always users of LCGPs. Individuals were classified based on insurance type (private, Medicare, Medicaid, dual-eligible). The proportion of days covered (PDC) was calculated for each medication class and the proportion of users with PDC ≥ 0.80 was reported as an observed metric for what would be calculated based on claims data and a true metric which included missing medication fills due to LCGPs. True measures of adherence were higher than the observed measures. The effect’s magnitude was highest for private insurance and for medication classes utilized more often through LCGPs. Thus, medication-based quality measures may be underestimated due to LCGPs.