Development and evaluation of an algorithm to link mothers and infants in two US commercial healthcare claims databases for pharmacoepidemiology research.

BACKGROUND: Administrative healthcare claims databases are used in drug safety research but are limited for investigating the impacts of prenatal exposures on neonatal and pediatric outcomes without mother-infant pair identification. Further, existing algorithms are not transportable across data sources. We developed a transportable mother-infant linkage algorithm and evaluated it in two, large US commercially insured populations. METHODS: We used two US commercial health insurance claims databases during the years 2000 to 2021. Mother-infant links were constructed where persons of female sex 12-55 years of age with a pregnancy episode ending in live birth were associated with a person who was 0 years of age at database entry, who shared a common insurance plan ID, had overlapping insurance coverage time, and whose date of birth was within ± 60-days of the mother's pregnancy episode live birth date. We compared the characteristics of linked vs. non-linked mothers and infants to assess similarity. RESULTS: The algorithm linked 3,477,960 mothers to 4,160,284 infants in the two databases. Linked mothers and linked infants comprised 73.6% of all mothers and 49.1% of all infants, respectively. 94.9% of linked infants' dates of birth were within ± 30-days of the associated mother's pregnancy episode end dates. Characteristics were largely similar in linked vs. non-linked mothers and infants. Differences included that linked mothers were older, had longer pregnancy episodes, and had greater post-pregnancy observation time than mothers with live births who were not linked. Linked infants had less observation time and greater healthcare utilization than non-linked infants. CONCLUSIONS: We developed a mother-infant linkage algorithm and applied it to two US commercial healthcare claims databases that achieved a high linkage proportion and demonstrated that linked and non-linked mother and infant cohorts were similar. Transparent, reusable algorithms applied to large databases enable large-scale research on exposures during pregnancy and pediatric outcomes with relevance to drug safety. These features suggest studies using this algorithm can produce valid and generalizable evidence to inform clinical, policy, and regulatory decisions.

Live birth prevalence of hemolytic disease of the fetus and newborn in the United States from 1996 to 2010.

BACKGROUND: Hemolytic disease of the fetus and newborn (HDFN) is mediated by maternal alloantibodies, a consequence of immune sensitization during pregnancy with maternal-fetal incompatibility with ABO, Rhesus factor (Rh), and/or other red blood cell antigens. RhD, Kell, and other non-ABO alloantibodies are the primary cause of moderate to severe HDFN, whereas ABO HDFN is typically mild. HDFN live birth prevalence owing to Rh alloimmunization among newborns in the United States was last estimated to be 106 per 100,000 births in 1986. HDFN live birth prevalence owing to all alloantibodies was estimated to be 817 to 840 per 100,000 in Europe. There is a need for updated prevalence estimates in the United States and a better understanding of disease demographics, severity, and treatments. OBJECTIVE: This study aimed to estimate the live birth prevalence of HDFN and the proportion of severe cases of HDFN in the United States, to describe the associated risk factors, and to compare the clinical outcomes and treatments among healthy newborns, newborns with HDFN, and newborns who are sick without HDFN using a nationally representative hospital discharge database. STUDY DESIGN: In this retrospective, observational cohort study, we used data from the 1996 to 2010 National Hospital Discharge Survey to identify live births, defined by inpatient visits with the newborn flag, with and without a diagnosis of HDFN across 200 to 500 sampled hospitals (≥6 beds) per year. Patient and hospital characteristics, alloimmunization status, disease severity, treatment, and clinical outcomes were evaluated. Frequencies and weighted percentages were calculated for all variables. Logistic regression was used to compare the characteristics between newborns with HDFN and other newborns using odds ratios. RESULTS: Of 480,245 live births identified, 9810 HDFN cases were recorded. When weighted to the United States population, this corresponded to a live birth prevalence of 1695 per 100,000 live births. Compared with other newborns, newborns with HDFN were more likely to be female, Black, living in the South (vs the Midwest or West), and treated at larger (>100 beds) and government-owned hospitals. ABO and Rh alloimmunization accounted for 78.1% and 4.3% of newborns with HDFN, respectively, whereas HDFN caused by other antigens, such as Kell and Duffy, accounted for 17.6% of the cases. Among newborns with HDFN, 22% received phototherapy, 1% received simple transfusions, and 0.5% received exchange transfusions or intravenous immunoglobulin. Newborns affected by HDFN caused by Rh alloimmunization were more likely to require medical interventions, including simple or exchange transfusions, and more likely to be delivered by cesarean delivery. Overall, HDFN was associated with a longer hospital length of stay in the neonatal intensive care unit when compared with healthy and other sick newborns, a higher rate of cesarean delivery, and a higher rate of nonroutine discharge than healthy newborns. CONCLUSION: Overall, the live birth prevalence of HDFN was higher than those previously reported, whereas Rh-induced HDFN live birth prevalence was similar to those previously reported. HDFN live birth prevalence owing to Rh alloimmunization decreased over time, likely because of continued Rh immune globulin prophylaxis. Treatment patterns for newborns with HDFN and the comparative clinical outcomes when compared with healthy newborns confirm the continued clinical needs of this population.

A concept-wide association study to identify potential risk factors for nonadherence among prevalent users of antihypertensives.

PURPOSE: We sought to determine whether an association study using information contained in clinical notes could identify known and potentially novel risk factors for nonadherence to antihypertensive medications. METHODS: We conducted a retrospective concept-wide association study (CWAS) using clinical notes to identify potential risk factors for medication nonadherence, adjusting for age, sex, race, baseline blood pressure, estimated glomerular filtration rate, and a combined comorbidity score. Participants included Medicare beneficiaries 65 years and older receiving care at the Harvard Vanguard Medical Associates network from 2010-2012 and enrolled in a Medicare Advantage program. Concepts were extracted from clinical notes in the year prior to the index prescription date for each patient. We tested associations with the outcome for 5013 concepts extracted from clinical notes in a derivation cohort (4382 patients) and accounted for multiple hypothesis testing by using a false discovery rate threshold of less than 5% (q < .05). We then confirmed the associations in a validation cohort (3836 patients). Medication nonadherence was defined using a proportion of days covered (PDC) threshold less than 0.8 using pharmacy claims data. RESULTS: We found 415 concepts associated with nonadherence, which we organized into 11 clusters using a hierarchical clustering approach. Volume depletion and overload, assessment of needs at the point of discharge, mood disorders, neurological disorders, complex coordination of care, and documentation of noncompliance were some of the factors associated with nonadherence. CONCLUSIONS: This approach was successful in identifying previously described and potentially new risk factors for antihypertensive nonadherence using the clinical narrative.

Effectiveness of Targeted Insulin-Adherence Interventions for Glycemic Control Using Predictive Analytics Among Patients With Type 2 Diabetes: A Randomized Clinical Trial.

Importance: Patient adherence to antidiabetic medications, especially insulin, remains poor, leading to adverse outcomes and increased costs. Most adherence interventions have only been modestly effective, partly because they are not targeted to patients who could benefit most. Objective: To evaluate whether delivering more intensive insulin-adherence interventions only to individuals with type 2 diabetes predicted to benefit most was more effective than delivering a lower-intensity intervention to a larger group of unselected individuals. Design, Setting, and Participants: This 3-arm pragmatic randomized clinical trial used data from Horizon, the largest health insurer in New Jersey, on 6000 participants 18 years or older with type 2 diabetes who were receiving basal insulin. Patients were excluded if they were insured by Medicaid or Medicare or had fewer than 3 months of continuous enrollment. The study was conducted from July 7, 2016, through October 5, 2017. Analyses were conducted from February 5 to September 24, 2018. Interventions: Eligible patients were randomized to 3 arms in a 1:1:1 ratio. Randomization was stratified based on baseline availability of 1 or more glycated hemoglobin A1c (HbA1c) test values. All arms were designed to cost the same, and each cohort received a tailored pharmacist telephone consultation varying based on (1) proportion receiving the intervention and (2) intensity, including follow-up frequency and cointerventions. Arm 1 offered a low-intensity intervention to all patients. Arm 2 offered a moderate-intensity intervention to 60% of patients based on their predicted risk of insulin nonadherence. Arm 3 offered a high-intensity intervention to 40% of patients based on glycemic control and predicted risk of insulin nonadherence. Main Outcomes and Measures: The primary outcome was insulin persistence. Secondary outcomes were changes in HbA1c level and health care utilization. Outcomes were evaluated in arms 2 and 3 vs arm 1 using claims data, intention-to-treat principles, and multiple imputation for missing values in the 12-month follow-up. Results: Among 6000 participants, mean (SD) age was 55.9 (11.0) years and 3344 (59.8%) were male. Compared with arm 1, insulin nonpersistence did not differ in arm 2 (relative risk, 0.88; 95% CI, 0.75-1.03) or arm 3 (relative risk, 0.91; 95% CI, 0.77-1.06). Glycemic control was similar in arm 2 and arm 1 (absolute HbA1c level difference, -0.15%; 95% CI, -0.34% to 0.05%) but was better in arm 3 (absolute HbA1c level difference, -0.25%; 95% CI, -0.43% to -0.06%). Total spending and office visits did not differ, but arm 2 (moderate intensity intervention) had more hospitalizations (odds ratio, 1.22; 95% CI, 1.06-1.41) and emergency department visits (odds ratio, 1.38; 95% CI, 1.24-1.53) than did arm 1 (low intensity intervention). Conclusions and Relevance: Compared with an untargeted low-intensity intervention, delivering a highly targeted high-intensity intervention did not improve insulin persistence but modestly improved mean glycemic control. A partially targeted moderate-intensity intervention did not change insulin persistence or HbA1c level but was associated with a small increase in hospitalizations. Trial Registration: ClinicalTrials.gov Identifier: NCT02846779.

Using Previous Medication Adherence to Predict Future Adherence.

BACKGROUND: Medication nonadherence is a major public health problem. Identification of patients who are likely to be and not be adherent can guide targeted interventions and improve the design of comparative-effectiveness studies. OBJECTIVE: To evaluate multiple measures of patient previous medication adherence in light of predicting future statin adherence in a large U.S. administrative claims database. METHODS: We identified a cohort of patients newly initiating statins and measured their previous adherence to other chronic preventive medications during a 365-day baseline period, using metrics such as proportion of days covered (PDC), lack of second fills, and number of dispensations. We measured adherence to statins during the year after initiation, defining high adherence as PDC ≥ 80%. We built logistic regression models from different combinations of baseline variables and previous adherence measures to predict high adherence in a random 50% sample and tested their discrimination using concordance statistics (c-statistics) in the other 50%. We also assessed the association between previous adherence and subsequent statin high adherence by fitting a modified Poisson model from all relevant covariates plus previous mean PDC categorized as < 25%, 25%-79%, and ≥ 80%. RESULTS: Among 89,490 statin initiators identified, a prediction model including only demographic variables had a c-statistic of 0.578 (95% CI = 0.573-0.584). A model combining information on patient comorbidities, health care services utilization, and medication use resulted in a c-statistic of 0.665 (95% CI = 0.659-0.670). Models with each of the previous medication adherence measures as the only explanatory variable yielded c-statistics ranging between 0.533 (95% CI = 0.529-0.537) for lack of second fill and 0.666 (95% CI = 0.661-0.671) for maximum PDC. Adding mean PDC to the combined model yielded a c-statistic of 0.695 (95% CI = 0.690-0.700). Given a sensitivity of 75%, the predictor improved the specificity from 47.7% to 53.6%. Patients with previous mean PDC < 25% were half as likely to show high adherence to statins compared with those with previous mean PDC ≥ 80% (risk ratio = 0.49, 95% CI = 0.46-0.50). CONCLUSIONS: Including measures of previous medication adherence yields better prediction of future statin adherence than usual baseline clinical measures that are typically used in claims-based studies. DISCLOSURES: This study was funded by the Patient-Centered Outcomes Research Institute (ME-1309-06274). Kumamaru, Kohsaka, and Miyata are affiliated with the Department of Healthcare Quality Assessment at the University of Tokyo, which is a social collaboration department supported by National Clinical Database. The department was formerly supported by endowments from Johnson & Johnson K.K., Nipro, Teijin Pharma, Kaketsuken K.K., St. Jude Medical Japan, Novartis Pharma K.K., Taiho Pharmaceutical, W. L. Gore & Associates, Olympus Corporation, and Chugai Pharmaceutical. Gagne has received grants from Novartis Pharmaceuticals and Eli Lilly and Company to the Brigham and Women's Hospital for unrelated work. He is a consultant to Aetion, a software company, and to Optum. Choudhry has received grants from the National Heart, Lung, and Blood Institute, PhRMA Foundation, Merck, Sanofi, AstraZeneca, CVS, and MediSafe. Schneeweiss is consultant to WHISCON and Aetion, a software manufacturer of which he also owns equity. He is principal investigator of investigator-initiated grants to the Brigham and Women's Hospital from Bayer, Genentech, and Boehringer Ingelheim unrelated to the topic of this study. He does not receive personal fees from biopharmaceutical companies. No potential conflict of interest was reported by the other authors.

Time to Filling of New Prescriptions for Chronic Disease Medications Among a Cohort of Elderly Patients in the USA.

BACKGROUND: Data on primary nonadherence remains sparse, due to a lack of data resources that combine information on medication prescribing and dispensing. In addition, previous work on primary nonadherence has used follow-up periods ranging from 30 days up to 18 months, making results difficult to compare. OBJECTIVE: To evaluate the prevalence and predictors of primary nonadherence by measuring time until filling in a cohort of elderly patients. DESIGN: Retrospective cohort study of new prescription episodes. PATIENTS: Data comes from a linked database of electronic health records and claims for patients aged ≥ 65 years enrolled in Medicare Parts A, B, and D during 2007-2014. We identified patients receiving a new prescription for a chronic disease medication with continuous Medicare enrollment for 180 days prior to the index prescription order and no fills or orders for the medication during this period. MAIN MEASURES: Time until filling of the index prescription for up to 1 year. KEY RESULTS: In 32,586 new medication orders, the majority (75%; 95% confidence interval [CI] 74-75%) of new prescriptions were filled within 7 days, 81% (81-82%) were filled within 30 days, and 91% (91-92%) were filled within 1 year. The rate and timing of dispensing were similar across therapeutic areas. Timing of initial filling within 7 days or within 30 days could be predicted with moderate accuracy (C-statistics = 0.70-0.74). Patients with > 5 current medications on hand at the time of the index prescription and average out-of-pocket medication costs < $5 filled 89% of prescriptions within 7 days. Patients with no current medications and out-of-pocket costs > $50 filled only 25% of prescriptions within 7 days. CONCLUSIONS: Nearly 20% of patients do not fill a new chronic disease prescription within 30 days. Patients with fewer recent fills and higher out-of-pocket costs are at higher risk of primary nonadherence.

Defining Exposure in Observational Studies Comparing Outcomes of Treatment Discontinuation.

BACKGROUND: Continuation of antiplatelet therapy beyond 12 months after a drug-eluting stent procedure reduced the risk of a major adverse cardiovascular and cerebrovascular event (MACCE) in the DAPT trial (Dual Antiplatelet Therapy). Observational studies have evaluated outcomes related to different durations of therapy but are susceptible to bias. METHODS AND RESULTS: Using deidentified claims from commercially insured and Medicare populations in the United States, we compared how increasingly stringent definitions of exposure affect associations between antiplatelet continuation versus discontinuation and MACCE, myocardial infarction, and intracerebral hemorrhage or gastrointestinal bleeding in patients meeting DAPT trial inclusion criteria between 2004 and 2013. Therapy continuation at 12 months was defined as (1) having antiplatelet supply on hand versus not (landmark time); (2) refilling within 30 days versus not among individuals with antiplatelet supply; (3) criteria 2 plus continuous prior antiplatelet use; and (4) criteria 2 and 3 plus a cardiologist visit in months 10 to 12. Propensity score-adjusted hazard ratios were compared. Cohort sizes were 53 679, 27 524, 16 971, and 7948, respectively, of which 20% were discontinuers on average. Increasing restriction led to progressively larger associations with continued treatment: cohort 1 MACCE hazard ratio, 0.79 (0.73, 0.87); myocardial infarction, 0.74 (0.65, 0.83); bleed, 1.03 (0.96, 1.11) versus cohort 4 MACCE hazard ratio, 0.66 (0.48, 0.91); myocardial infarction, 0.56 (0.37, 0.86); bleed, 1.24 (0.95, 1.61). Estimates trended toward DAPT trial estimates and were associated with reduced levels of exposure misclassification. CONCLUSIONS: In an example of long-term antiplatelet use, increasing restrictions on the definition of therapy continuation yielded results consistent with trial estimates by reducing exposure misclassification.

Predicting Adherence to Chronic Disease Medications in Patients with Long-term Initial Medication Fills Using Indicators of Clinical Events and Health Behaviors.

BACKGROUND: Efforts at predicting long-term adherence to medications have been focused on patients filling typical month-long supplies of medication. However, prediction remains difficult for patients filling longer initial supplies, a practice that is becoming increasingly common as a method to enhance medication adherence. OBJECTIVES: To (a) extend methods involving short-term filling behaviors and (b) develop novel variables to predict adherence in a cohort of patients receiving longer initial prescriptions. METHODS: In this retrospective cohort study, we used claims from a large national insurer to identify patients initiating a 90-day supply of oral medications for diabetes, hypertension, and hyperlipidemia (i.e., statins). Patients were included in the cohort if they had continuous database enrollment in the 180 days before and 365 days after medication initiation. Adherence was measured in the subsequent 12 months using the proportion of days covered metric. In total, 125 demographic, clinical, and medication characteristics at baseline and in the first 30-120 days after initiation were used to predict adherence using logistic regression models. We used 10-fold cross-validation to assess predictive accuracy by discrimination (c-statistic) measures. RESULTS: In total, 32,249 patients met the inclusion criteria, including 14,930 patients initiating statins, 12,887 patients initiating antihypertensives, and 4,432 patients initiating oral hypoglycemics. Prediction using only baseline variables was relatively poor (cross-validated c-statistic = 0.644). Including indicators of acute clinical conditions, health resource utilization, and short-term medication filling in the first 120 days greatly improved predictive ability (0.823). A model that incorporated all baseline characteristics and predictors within the first 120 days after medication initiation more accurately predicted future adherence (0.832). The best performing model that included all 125 baseline and postbaseline characteristics had strong predictive ability (0.837), suggesting the utility of measuring these novel postbaseline variables in this population. CONCLUSIONS: We demonstrate that long-term, 12-month adherence in patients filling longer supplies of medication can be strongly predicted using a combination of clinical, health resource utilization, and medication filling characteristics before and after treatment initiation. DISCLOSURES: This work was supported by an unrestricted grant from CVS Health to Brigham and Women's Hospital. Shrank and Matlin were employees and shareholders at CVS Health at the time of this study; they report no financial interests in products or services that are related to this subject. Spettell is an employee of, and shareholder in, Aetna. This research was previously presented at the 2016 Annual Conference of the International Society for Pharmacoepidemiology; August 25-28, 2016; Dublin, Ireland.

The relative benefits of claims and electronic health record data for predicting medication adherence trajectory.

BACKGROUND: Healthcare providers are increasingly encouraged to improve their patients' adherence to chronic disease medications. Prediction of adherence can identify patients in need of intervention, but most prediction efforts have focused on claims data, which may be unavailable to providers. Electronic health records (EHR) are readily available and may provide richer information with which to predict adherence than is currently available through claims. METHODS: In a linked database of complete Medicare Advantage claims and comprehensive EHR from a multi-specialty outpatient practice, we identified patients who filled a prescription for a statin, antihypertensive, or oral antidiabetic during 2011 to 2012. We followed patients to identify subsequent medication filling patterns and used group-based trajectory models to assign patients to adherence trajectories. We then identified potential predictors from both claims and EHR data and fit a series of models to evaluate the accuracy of each data source in predicting medication adherence. RESULTS: Claims were highly predictive of patients in the worst adherence trajectory (C=0.78), but EHR data also provided good predictions (C=0.72). Among claims predictors, presence of a prior gap in filling of at least 6 days was by far the most influential predictor. In contrast, good predictions from EHR data required complex models with many variables. CONCLUSION: EHR data can provide good predictions of adherence trajectory and therefore may be useful for providers seeking to deploy resource-intensive interventions. However, prior adherence information derived from claims is most predictive, and can supplement EHR data when it is available.

Medication Synchronization Programs Improve Adherence To Cardiovascular Medications And Health Care Use.

Medication synchronization programs based in pharmacies simplify the refill process by enabling patients to pick up all of their medications on a single visit. This can be especially important for improving medication adherence in patients with complex chronic diseases. We evaluated the impact of two synchronization programs on adherence, cardiovascular events, and resource use among Medicare beneficiaries treated between 2011 and 2014 for two or more chronic conditions-at least one of which was hypertension, hyperlipidemia, or diabetes. Among nearly 23,000 patients matched by propensity score, the mean proportion of days covered (a measure of medication adherence) for the control group of patients without a synchronization program was 0.84 compared to 0.87 for synchronized patients-a gain of 3 percentage points. Adherence improvement in synchronized versus control patients was three times greater in patients with low baseline adherence, compared to those with higher baseline adherence. Rates of hospitalization and emergency department visits and rates of outpatient visits were 9 percent and 3 percent lower in the synchronized group compared to the control group, respectively, while cardiovascular event rates were similar. Synchronization programs were associated with improved adherence for patients with cardiovascular disease, especially those with low baseline adherence.

Study protocol for the dabigatran, apixaban, rivaroxaban, edoxaban, warfarin comparative effectiveness research study.

Since 2010, four oral anticoagulants have been approved for marketing in addition to warfarin for treatment of thromboembolic disease. Limited head-to-head data exist comparing these treatments, leaving patients and clinicians with little guidance for selecting a strategy that balances recurrence reduction with bleeding risk. In the dabigatran, apixaban, rivaroxban, edoxaban and warfarin comparative effectiveness research study, we compare all five currently available oral anticoagulant agents for the extended treatment of deep venous thrombosis and pulmonary embolism, as well as no extended treatment, and evaluate whether results differ in specific sub-populations. As our population includes Medicare novel anticoagulant users and large numbers of commercially insured and Medicaid patients, our results will likely be transportable to the majority of US patients experiencing a DVT or pulmonary embolism. CLINICAL TRIALS REGISTRATION: NCT03271450.

The Impact of Text Messaging on Medication Adherence and Exercise Among Postmyocardial Infarction Patients: Randomized Controlled Pilot Trial.

BACKGROUND: Adherence to evidence-based therapies such as medications and exercise remains poor among patients after a myocardial infarction (MI). Text message reminders have been shown to improve rates of adherence to medication and exercise, but the existing studies have been of short duration. OBJECTIVE: Two single-center randomized controlled pilot trials were conducted to evaluate the impact of text message reminders over 12 months on adherence to cardiac medications and exercise among patients receiving cardiac rehabilitation after hospitalization for MI. METHODS: In the medication adherence trial, 34 patients were randomized to receive usual care alone or usual care plus daily text message reminders delivered at the time of day at which medications were to be taken. In the exercise adherence trial, 50 patients were randomized to receive usual care alone or usual care plus 4 daily text messages reminding them to exercise as directed. RESULTS: The text message reminders led to a mean 14.2 percentage point improvement in self-reported medication adherence over usual care (P<.001, 95% CI 7-21). In the exercise trial, text message reminders resulted in an additional 4.2 days (P=.001, 95% CI 1.9-6.4) and 4.0 hours (P<.001, 95% CI 2.4-5.6) of exercise per month over usual care and a nonsignificant increase of 1.2 metabolic equivalents (METS; P=.06) in exercise capacity as assessed by a BRUCE protocol at 12 months. CONCLUSIONS: Text message reminders significantly increased adherence to medication and exercise among post-MI patients receiving care in a structured cardiac rehabilitation program. This technology represents a simple and scalable method to ensure consistent use of evidence-based cardiovascular therapies. TRIAL REGISTRATION: Clinicaltrials.gov NCT02783287; https://clinicaltrials.gov/ct2/show/NCT02783287 (Archived by WebCite at http://www.webcitation.org/6sBnvNb05).

Predicting 1-Year Statin Adherence Among Prevalent Users: A Retrospective Cohort Study.

BACKGROUND: Attempts to predict who is at risk of future nonadherence have largely focused on predictions at the time of therapy initiation; however, these users are only a small proportion of all patients on therapy at any point in time. Methods to predict nonadherence for established medication users, which have not been previously described in the literature, would be helpful to guide efforts to enhance the use of evidence-based therapies. OBJECTIVE: To test approaches for adherence prediction among prevalent statin users, namely the use of short-term filling behavior, investigator-specified predictors from medical and pharmacy administrative claims, and the empirical selection of potential predictors using the high-dimensional propensity score variable selection algorithm. METHODS: Medical and prescription claims data from a large national health insurer were used to create a cohort of patients who filled statin medication prescriptions in January 2012. We defined 6 groups of adherence predictors and estimated 10 main models to predict medication adherence in the full cohort. The same was done for the population stratified based on the days supply of the index statin prescription (≤ 30 days vs. > 30 days). RESULTS: The study cohort consisted of 93,777 individuals, 58.4% of which were adherent to statins during follow-up. The use of 3 pre-index adherence predictors alone achieved a c-statistic of 0.70. Investigator-specified and empirically selected pharmacy, medical, and demographic variables did substantially worse (0.57-0.60). The use of 3 indicators of post-index adherence achieved a higher c-statistic than the best-performing model using pre-index information (0.74 vs. 0.72). The addition of 3 pre-index adherence predictors further improved discrimination (0.78). CONCLUSIONS: This analysis demonstrated the ability to predict adherence among medication users using filling behavior before and immediately after an index prescription fill. DISCLOSURES: This work was supported by an unrestricted grant from CVS Health to Brigham and Women's Hospital. Shrank, Brennan, and Matlin were employees and shareholders at CVS Health at the time of this manuscript preparation; they report no financial interests in products or services that are related to the subject of the manuscript. Franklin has received consulting fees from Aetion. Chourdry has received grants from the National Heart, Lung, and Blood Institute, PhRMA Foundation, Merck, Sanofi, AstraZeneca, and MediSafe. Spettell is an employee of, and shareholder in, Aetna. The other authors have nothing to disclose. Krumme, Choudhry, Tong, and Franklin contributed to the study design, interpretation of results, and manuscript drafting. Tong prepared and analyzed the data. Isaman, Spettell, Shrank, Brennan, and Matlin provided interpretation of results and critical manuscript revisions.

Effect of Reminder Devices on Medication Adherence: The REMIND Randomized Clinical Trial.

Importance: Forgetfulness is a major contributor to nonadherence to chronic disease medications and could be addressed with medication reminder devices. Objective: To compare the effect of 3 low-cost reminder devices on medication adherence. Design, Setting, and Participants: This 4-arm, block-randomized clinical trial involved 53 480 enrollees of CVS Caremark, a pharmacy benefit manager, across the United States. Eligible participants were aged 18 to 64 years and taking 1 to 3 oral medications for long-term use. Participants had to be suboptimally adherent to all of their prescribed therapies (with a medication possession ratio of 30% to 80%) in the 12 months before randomization. Participants were stratified on the basis of the medications they were using at randomization: medications for cardiovascular or other nondepression chronic conditions (the chronic disease stratum) and antidepressants (the antidepressant stratum). In each stratum, randomization occurred within blocks defined by whether all of the patient's targeted medications were dosed once daily. Patients were randomized to receive in the mail a pill bottle strip with toggles, digital timer cap, or standard pillbox. The control group received neither notification nor a device. Data were collected from February 12, 2013, through March 21, 2015, and data analyses were on the intention-to-treat population. Main Outcomes and Measures: The primary outcome was optimal adherence (medication possession ratio ≥80%) to all eligible medications among patients in the chronic disease stratum during 12 months of follow-up, ascertained using pharmacy claims data. Secondary outcomes included optimal adherence to cardiovascular medications among patients in the chronic disease stratum as well as optimal adherence to antidepressants. Results: Of the 53 480 participants, mean (SD) age was 45 (12) years and 56% were female. In the primary analysis, 15.5% of patients in the chronic disease stratum assigned to the standard pillbox, 15.1% assigned to the digital timer cap, 16.3% assigned to the pill bottle strip with toggles, and 15.1% assigned to the control arm were optimally adherent to their prescribed treatments during follow-up. There was no statistically significant difference in the odds of optimal adherence between the control and any of the devices (standard pillbox: odds ratio [OR], 1.03 [95% CI, 0.95-1.13]; digital timer cap: OR, 1.00 [95% CI, 0.92-1.09]; and pill bottle strip with toggles: OR, 0.94 [95% CI, 0.85-1.04]). In direct comparisons, the odds of optimal adherence were higher with a standard pillbox than with the pill bottle strip (OR, 1.10 [95% CI, 1.00-1.21]). Secondary analyses yielded similar results. Conclusions and Relevance: Low-cost reminder devices did not improve adherence among nonadherent patients who were taking up to 3 medications to treat common chronic conditions. The devices may have been more effective if coupled with interventions to ensure consistent use or if targeted to individuals with an even higher risk of nonadherence. Trial Registration: clinicaltrials.gov Identifier: NCT02015806.

Longitudinal Patterns of Spending Enhance the Ability to Predict Costly Patients: A Novel Approach to Identify Patients for Cost Containment.

BACKGROUND: With rising health spending, predicting costs is essential to identify patients for interventions. Many of the existing approaches have moderate predictive ability, which may result, in part, from not considering potentially meaningful changes in spending over time. Group-based trajectory modeling could be used to classify patients into dynamic long-term spending patterns. OBJECTIVES: To classify patients by their spending patterns over a 1-year period and to assess the ability of models to predict patients in the highest spending trajectory and the top 5% of annual spending using prior-year predictors. SUBJECTS: We identified all fully insured adult members enrolled in a large US nationwide insurer and used medical and prescription data from 2009 to 2011. RESEARCH DESIGN: Group-based trajectory modeling was used to classify patients by their spending patterns over a 1-year period. We assessed the predictive ability of models that categorized patients in the top fifth percentile of annual spending and in the highest spending trajectory, using logistic regression and split-sample validation. Models were estimated using investigator-specified variables and a proprietary risk-adjustment method. RESULTS: Among 998,651 patients, in the best-performing model, prediction was strong for patients in the highest trajectory group (C-statistic: 0.86; R: 0.47). The C-statistic of being in the top fifth percentile of spending in the best-performing model was 0.82 (R: 0.26). Approaches using nonproprietary investigator-specified methods performed almost as well as other risk-adjustment methods (C-statistic: 0.81 vs. 0.82). CONCLUSIONS: Trajectory modeling may be a useful way to predict costly patients that could be implementable by payers to improve cost-containment efforts.

Association Between Patient-Centered Medical Homes and Adherence to Chronic Disease Medications: A Cohort Study.

BACKGROUND: Despite the widespread adoption of patient-centered medical homes into primary care practice, the evidence supporting their effect on health care outcomes has come primarily from geographically localized and well-integrated health systems. OBJECTIVE: To assess the association between medication adherence and medical homes in a national patient and provider population, given the strong ties between adherence to chronic disease medications and health care quality and spending. DESIGN: Retrospective cohort study. SETTING: Claims from a large national health insurer. PATIENTS: Patients initiating therapy with common medications for chronic diseases (diabetes, hypertension, and hyperlipidemia) between 2011 and 2013. MEASUREMENTS: Medication adherence in the 12 months after treatment initiation was compared among patients cared for by providers practicing in National Committee for Quality Assurance-recognized patient-centered medical homes and propensity score-matched control practices in the same Primary Care Service Areas. Linear mixed models were used to examine the association between medical homes and adherence. RESULTS: Of 313 765 patients meeting study criteria, 18 611 (5.9%) received care in patient-centered medical homes. Mean rates of adherence were 64% among medical home patients and 59% among control patients. Among 4660 matched control and medical home practices, medication adherence was significantly higher in medical homes (2.2% [95% CI, 1.5% to 2.9%]). The association between medical homes and better adherence did not differ significantly by disease state (diabetes, 3.0% [CI, 1.5% to 4.6%]; hypertension, 3.2% [CI, 2.2% to 4.2%]; hyperlipidemia, 1.5% [CI, 0.6% to 2.5%]). LIMITATION: Clinical outcomes related to medication adherence were not assessed. CONCLUSION: Receipt of care in a patient-centered medical home is associated with better adherence, a vital measure of health care quality, among patients initiating treatment with medications for common high-cost chronic diseases. PRIMARY FUNDING SOURCE: CVS Health.

Prevalence, effectiveness, and characteristics of pharmacy-based medication synchronization programs.

OBJECTIVES: The burden of visiting pharmacies to fill medications is a central contributor to nonadherence to maintenance medications. Recently, pharmacies have begun offering services that align prescription fill dates to allow patients to pick up all medications on a single visit. We evaluated the prevalence and structure of synchronization programs and evidence of their impact on adherence and clinical outcomes. STUDY DESIGN: Mixed-methods approach consisting of semi-structured interviews, data from surveillance activities, and a systematic literature review. METHODS: We conducted interviews with opinion leaders from nonprofit advocacy organizations and exemplary synchronization programs. Program prevalence was determined using data from regular surveillance efforts. A literature review included Medline, EMBASE, Google Scholar, and general Internet searches. RESULTS: Synchronization programs exist in approximately 10% of independent, 6% of stand-alone chain, and 11% of retail store pharmacies. The majority of programs include a monthly pharmacist appointment and reminder communication. Programs reported the importance of pharmacist buy-in, technology to track and recruit patients, links to other healthcare services, and flexible solutions for managing costs and communication preferences. Although existing peer-reviewed literature suggests that synchronization improves adherence, more evidence is needed to evaluate its impact on patient-centered outcomes. CONCLUSIONS: As medication synchronization programs shift directions and compete for patients and payer resources, it will be more important than ever to rigorously evaluate their ability to improve clinical outcomes while also providing the growing number of patients managing multiple chronic conditions with the highest level of patient engagement and consumer choice.

Brief Report: Intensification to Triple Therapy After Treatment With Nonbiologic Disease-Modifying Antirheumatic Drugs for Rheumatoid Arthritis in the United States From 2009 to 2014.

OBJECTIVE: Several trials suggest that triple therapy (methotrexate, sulfasalazine, and hydroxychloroquine) and biologic disease-modifying antirheumatic drugs (DMARDs) have similar efficacy in patients with rheumatoid arthritis (RA). This study was undertaken to investigate intensification to triple therapy after initial nonbiologic prescription among patients with RA. METHODS: The use of triple therapy among patients with RA in 2009-2014 was evaluated using US insurance claims data. Patients with a health care visit for RA and an initial nonbiologic DMARD prescription were included. Frequencies of intensification to triple therapy or a biologic DMARD and rates of intensification per 6-month time period were calculated. Using Cox regression, we evaluated whether sociodemographic, temporal, geographic, clinical, and health care utilization factors were associated with intensification to triple therapy. Among those patients whose therapy was intensified, we investigated factors associated with triple therapy use by logistic regression. Hazard ratios (HRs) with 95% confidence intervals (95% CIs) for intensification to triple therapy in relation to various clinical and demographic factors were calculated. RESULTS: There were 24,576 patients with a mean ± SD age of 50.3 ± 12.3 years, and 78% were female. During the study period, treatment was intensified to biologic DMARDs in 2,739 patients (11.1%) compared to 181 patients (0.7%) whose treatment was intensified to triple therapy. There was no significant change in triple therapy use across calendar years. Patients whose treatment was intensified to triple therapy were more likely to receive glucocorticoids (HR 1.91 [95% CI 1.41-2.60]) compared to patients who did not use glucocorticoids and were more likely to use nonsteroidal antiinflammatory drugs (NSAIDs) (HR 1.48, 95% CI 1.10-1.99 versus no NSAID use). Among those patients whose treatment was intensified to triple therapy or biologic DMARDs, factors significantly associated with triple therapy use included older age, US region (with the highest odds for triple therapy use in the West and lowest odds for triple therapy use in the Northeast), glucocorticoid use, and lower number of outpatient visits within 180 days of initial nonbiologic DMARD prescription. CONCLUSION: Despite reports published during the study period suggesting equivalent efficacy of triple therapy and biologic DMARDs for RA, the use of triple therapy was infrequent and did not increase over time in this large nationwide study.

Can purchasing information be used to predict adherence to cardiovascular medications? An analysis of linked retail pharmacy and insurance claims data.

OBJECTIVE: The use of retail purchasing data may improve adherence prediction over approaches using healthcare insurance claims alone. DESIGN: Retrospective. SETTING AND PARTICIPANTS: A cohort of patients who received prescription medication benefits through CVS Caremark, used a CVS Pharmacy ExtraCare Health Care (ECHC) loyalty card, and initiated a statin medication in 2011. OUTCOME: We evaluated associations between retail purchasing patterns and optimal adherence to statins in the 12 subsequent months. RESULTS: Among 11 010 statin initiators, 43% were optimally adherent at 12 months of follow-up. Greater numbers of store visits per month and dollar amount per visit were positively associated with optimal adherence, as was making a purchase on the same day as filling a prescription (p<0.0001 for all). Models to predict adherence using retail purchase variables had low discriminative ability (C-statistic: 0.563), while models with both clinical and retail purchase variables achieved a C-statistic of 0.617. CONCLUSIONS: While the use of retail purchases may improve the discriminative ability of claims-based approaches, these data alone appear inadequate for adherence prediction, even with the addition of more complex analytical approaches. Nevertheless, associations between retail purchasing behaviours and adherence could inform the development of quality improvement interventions.