Increasing tramadol utilisation under strict regulatory control of opioid prescribing - A cross-sectional study in Taiwan from 2002 through 2016.

BACKGROUND/PURPOSE: Prescribing of opioids to patients with non-cancer pain is strictly regulated in Taiwan, but tramadol is not included in the regulation on chronic opioid prescribing. This study aims to identify the utilization trend of prescribing tramadol and other opioid analgesics and investigate the influence of government regulation on opioid prescribing in Taiwan. METHODS: This cross-sectional study used the Taiwan National Health Insurance claims database and the cancer registry from 2001 through 2016. The annual number of adult opioid users, opioid utilization (Defined Daily Doses [DDDs]/1000 registrants) and the number of supply days were enumerated for each calendar year and stratified by cancer or non-cancer patients. Descriptive statistics were used to report the trends in utilization for each calendar year. RESULTS: The regulation strictly limited persistent use of opioids for patients with non-cancer pain, of which only a small proportion of fentanyl (20%) and morphine (<2%) users were prescribed with an annual number of supply days greater than 28 days. The annual utilization of morphine (6.4-53.5 vs. 1.1 to 9.6 DDD/1000 registrants) and fentanyl (8.3-37.0 vs. 0.16 to 1.8 DDD/1000 registrants) to patients with cancer was consistently higher than patients without cancer. In contrast to morphine and fentanyl, the utilization of tramadol prescribed to patients without cancer increased 92.2-fold (3.7-341.2 DDD/1000 registrants) from 2002 to 2016. CONCLUSION: The regulation in Taiwan limited the prescribing of selective opioids for patients with non-cancer pain and the substitution of tramadol for other opioids may have safety implications.

Assessment of pre-specified adverse events following varicella vaccine: A population-based self-controlled risk interval study.

BACKGROUND: Clinical trials and spontaneous reporting systems have revealed rare but biologically plausible adverse events following varicella immunization. Few post-marketing controlled studies have been conducted to assess the relationship between the varicella vaccine and these outcomes. OBJECTIVES: To evaluate the risk of pneumonia, idiopathic thrombocytopenic purpura (ITP), meningitis, encephalitis and ischemic stroke following varicella immunization. MATERIALS AND METHODS: This nationwide observational study was based on Taiwan National Health Insurance data and National Immunization Information System from 2004 through 2014. Primary analysis included children aged 12-35 months who received the single varicella vaccine on the date of administration. The self-controlled risk interval design compared the incidence of pre-specified outcomes during a risk interval of 1-42 days post-vaccination and a control interval of 43-84 days. The outcomes of interest were defined as admitted pneumonia, ITP, meningitis, encephalitis, and ischemic stroke, as well as fracture as a negative control. Conditional Poisson regression was used to assess the incidence rate ratio (aIRR) with adjustments for age and seasonal effects. RESULTS: Among 1,194,189 children, who receiving the varicella vaccine, there was no observed increase in the risk for ITP (aIRR 1.00; 95% CI, 0.76-1.33), meningitis (aIRR 1.21; 95% CI, 0.49-2.95), encephalitis (aIRR 1.00; 95% CI, 0.62-1.60), or ischemic stroke (aIRR 1.24; 95% CI, 0.31-4.95). A clustering feature with pneumonia occurred during days 36-42 post-vaccination (aIRR 1.10; 95% CI, 1.02-1.18). An increase in the risk for ITP was observed in children receiving the varicella and MMR vaccines concomitantly (aIRR 1.70; 95% CI, 1.19-2.43), but not among those receiving the varicella vaccine only. CONCLUSIONS: We detected a small risk of incidental pneumonia associated with varicella vaccine in the 6th week after immunization. There was no increase in the risk of other pre-specified adverse events.

Burden of asthma and COPD overlap (ACO) in Taiwan: a nationwide population-based study.

BACKGROUND: Patients with symptoms of both asthma and chronic obstructive pulmonary disease (COPD) may be classified with the term asthma-COPD overlap (ACO). ACO is of considerable interest as it is currently poorly characterised and has been associated with worse health outcomes and higher healthcare costs compared with COPD or asthma alone. Patients with ACO in Asia remain poorly described, and there is limited information regarding their resource utilisation compared with patients with asthma or COPD only. This study investigated the characteristics, disease burden and medical resource utilisation of patients with ACO in Taiwan. METHODS: This was a retrospective cohort study of patients identified from National Health Insurance (NHI) claims data in Taiwan in 2009-2011. Patients were classified into incident ACO, COPD or asthma cohorts according to International Classification of Disease, ninth revision, clinical modification codes in claims. Eligible patients were ≥40 years of age with 12 months' continuous enrolment in the NHI programme pre- and post-index date (date of the first relevant medical claim). RESULTS: Patients with ACO (N = 22,328) and COPD (N = 69,648) were older and more likely to be male than those with asthma (N = 50,293). Patients with ACO had more comorbidities and exacerbations, with higher medication use: short-acting ß2-agonist prescriptions ranged from 30.4% of patients (asthma cohort) to 43.6% (ACO cohort), and inhaled corticosteroid/long-acting ß2-agonist combination prescriptions ranged from 11.1% (COPD cohort) to 35.0% (ACO cohort) in the 12 months following index. Patients with ACO generally had the highest medication costs of any cohort (long-acting muscarinic antagonist costs ranged from $227/patient [asthma cohort] to $349/patient [ACO cohort]); they also experienced more respiratory-related hospital visits than patients with asthma or COPD (mean outpatient/inpatient visits per patient post-index: 9.1/1.9 [ACO cohort] vs 5.7/1.4 [asthma cohort] and 6.4/1.7 [COPD cohort]). CONCLUSIONS: Patients with ACO in Taiwan experience a greater disease burden with greater healthcare resource utilisation, and higher costs, than patients with asthma or COPD alone.

Clinical outcomes in low risk coronary artery disease patients treated with different limus-based drug-eluting stents--a nationwide retrospective cohort study using insurance claims database.

The clinical outcomes of different limus-based drug-eluting stents (DES) in a real-world setting have not been well defined. The aim of this study was to investigate the clinical outcomes of three different limus-based DES, namely sirolimus-eluting stent (SES), Endeavor zotarolimus-eluting stent (E-ZES) and everolimus-eluting stent (EES), using a national insurance claims database. We identified all patients who received implantation of single SES, E-ZES or EES between January 1, 2007 and December 31, 2009 from the National Health Insurance claims database, Taiwan. Follow-up was through December 31, 2011 for all selected clinical outcomes. The primary end-point was all-cause mortality. Secondary end-points included acute coronary events, heart failure needing hospitalization, and cerebrovascular disease. Cox regression model adjusting for baseline characteristics was used to compare the relative risks of different outcomes among the three different limus-based DES. Totally, 6584 patients were evaluated (n=2142 for SES, n=3445 for E-ZES, and n=997 for EES). After adjusting for baseline characteristics, we found no statistically significant difference in the risk of all-cause mortality in three DES groups (adjusted hazard ratio [HR]: 1.14, 95% confidence interval [CI]: 0.94-1.38, p=0.20 in E-ZES group compared with SES group; adjusted HR: 0.77, 95% CI: 0.54-1.10, p=0.15 in EES group compared with SES group). Similarly, we found no difference in the three stent groups in risks of acute coronary events, heart failure needing hospitalization, and cerebrovascular disease. In conclusion, we observed no difference in all-cause mortality, acute coronary events, heart failure needing hospitalization, and cerebrovascular disease in patients treated with SES, E-ZES, and EES in a real-world population-based setting in Taiwan.

Impacts of drug reimbursement reductions on utilization and expenditures of oral antidiabetic medications in Taiwan: an interrupted time series study.

OBJECTIVES: To control increasing pharmaceutical expenditures, Taiwan's National Health Insurance has implemented a series of drug reimbursement price reductions since 2000. This study examined changes in use and expenditures of oral antidiabetic medications following the price regulation in November 2006. METHODS: We obtained claims data between January 2006 and August 2007 from Taiwan's National Health Insurance Research Database. We categorized oral antidiabetic products as affected by the reimbursement reduction ("targeted") or not ("non-targeted"), by level of relative price reduction, and by manufacturer type (international vs. local manufacturers). We used an interrupted time series design and segmented regression models to estimate changes in monthly per capita prescribing rate, volume, and insurance reimbursement expenditures following the policy. RESULTS: The majority (129/178; 72.5%) of oral antidiabetic products were targeted by this round of price reductions. There was a relative reduction of 9.5% [95%CI: -12.68, -6.32] in total expenditures at ten months post-policy compared to expected rates. For targeted products, there were 2.04% [95%CI: -4.15, 0.07] and 13.26% [95%CI: -16.64, -9.87] relative reductions in prescribing rate and expenditures, respectively, at ten months post-policy. Non-targeted products increased significantly (22% [95%CI: 10.49, 33.51] and 22.85% [95%CI: 11.69, 34.01] relative increases in prescribing rate and expenditures respectively). Larger reimbursement cuts led to greater reductions in prescribing rate, volume, and insurance reimbursement expenditures of targeted products. Prescribing rates of both targeted and non-targeted products by international manufacturers declined after the policy while rates of prescribing non-targeted products by local manufacturers increased. CONCLUSIONS: While total government expenditures for oral antidiabetic medications were contained by the policy, our results indicate that prescribing shifted at the margin from targeted to non-targeted products and from international to local products. Further research is warranted to understand how changes in medication use due to price regulation policies affect medication adherence and patient health outcomes.

Identification of seizures among adults and children following influenza vaccination using health insurance claims data.

INTRODUCTION: Post-licensure surveillance of adverse events following vaccination or prescription drug use often relies on electronic healthcare data to efficiently detect and evaluate safety signals. The accuracy of seizure-related diagnosis codes in identifying true incident seizure events in vaccine safety studies is influenced by factors such as clinical setting of diagnosis and age. To date, most studies of post-vaccination seizure have focused on pediatric populations. More information is needed on how well seizure can be identified in adults and children using algorithms that rely on electronic healthcare data. METHODS: This validation study was part of a larger safety study of influenza vaccination during the 2009-2010 and 2010-2011 influenza seasons. Children and adults receiving influenza vaccination were drawn from an administrative claims database of a large United States healthcare insurer. Potential seizure events were identified using an algorithm of ICD-9 diagnosis codes associated with an emergency department (ED) visit or hospitalization within pre-specified risk windows following influenza vaccination. Seizure events were confirmed through medical record review. The positive predictive value (PPV) of the algorithm was calculated within each diagnostic setting and stratified by age group, ICD-9 code group, and sex. RESULTS: Review confirmed 113 out of 176 potential seizure events. The PPVs were higher in the ED setting (93.9%) than in the inpatient setting (38.3%). The PPVs by age varied within the ED setting (98.2% in <7 years, 76.9% in 7-24 years, 92.3% in ≥25 years) and within the inpatient setting (64.7% in <7 years, 33.3% in 7-24 years, 32.3% in ≥25 years). CONCLUSIONS: Our algorithm for identification of seizure events using claims data had a high level of accuracy in the emergency department setting in young children and older adults and a lower, but acceptable, level of accuracy in older children and young adults.

Evaluating the safety of influenza vaccine using a claims-based health system.

INTRODUCTION: As part of the Centers for Disease Control and Prevention's monitoring and evaluation activities for influenza vaccines, we examined relationships between influenza vaccination and selected outcomes in the 2009-2010 and 2010-2011 influenza seasons in a claims-based data environment. METHODS: We included patients with claims for trivalent influenza vaccine (TIV) and/or 2009 pandemic influenza A H1N1 vaccine (H1N1) during the 2009-2010 and 2010-2011 influenza seasons. Patients were followed for several pre-specified outcomes identified in claims. Seizures and Guillain-Barré Syndrome were selected a priori for medical record confirmation. We estimated incidence rate ratios (IRR) using a self-controlled risk interval (SCRI) or a historical comparison design. Outcomes with elevated IRRs, not selected a priori for medical record review, were further investigated with review of claims histories surrounding the outcome date to determine whether the potential event could be ruled-out or attributed to other causes based on the pattern of medical care. RESULTS: In the 2009-2010 season, no significant increased risks for outcomes following H1N1 vaccination were observed. Following TIV administration, the IRR for peripheral nervous system disorders and neuropathy was slightly elevated (1.07, 95% CI: 1.01-1.13). The IRR for anaphylaxis following TIV was 28.55 (95% CI: 3.57-228.44). After further investigation of claims histories, the majority of potential anaphylaxis cases had additional claims around the time of the event indicating alternate explanatory factors or diagnoses. In the 2010-2011 season following TIV administration, a non-significant elevated IRR for anaphylaxis was observed with no other significant outcome findings. CONCLUSION: After claims history review, we ultimately found no increased outcome risk following administration of 998,881 TIV and 538,257 H1N1 vaccine doses in the 2009-2010 season, and 1,158,932 TIV doses in the 2010-2011 season.

Algorithms for identification of Guillain-Barré Syndrome among adolescents in claims databases.

BACKGROUND: Health insurance claims databases can provide data for studies of vaccine-related Guillain-Barre' Syndrome (GBS), but not all patients with a diagnostic ICD-9-CM code for GBS have the disease. The objective of this study was to evaluate the positive predictive values (PPVs) of claims-based algorithms for identifying GBS cases in 4 claims database environments. METHODS: Potential cases were adolescents ages 11-21 with at least one claim for GBS (ICD-9-CM code 357.0). Medical record reviews by a panel of 3 neurologists were conducted for case confirmation. Claims data considered for inclusion in the case-ascertainment algorithm included coding position, physician specialty, visit type, diagnostic tests. PPVs were used to assess the contribution of study factors in predicting case status. RESULTS: Among 361 individuals with a GBS diagnosis code, 106 were confirmed overall (PPV=0.29), varying from 0.24 to 0.56 across the 4 sites. Requiring the GBS code to be associated with a neurologist visit (PPV=0.53) or to be in a primary position on an inpatient claim (0.56) improved the performance. A composite algorithm including a primary inpatient GBS code and a neurologist visit associated with any GBS code gave the highest PPV (0.70). Incorporating claims for diagnostic testing had little impact on the PPV. Findings were generally similar across study sites. CONCLUSIONS: Algorithms were able to identify GBS cases better than the single occurrence of the diagnostic code for GBS, and these algorithms may perform similarly in different claims environments.

An economic evaluation of a laboratory monitoring program for renin-angiotensin system agents.

BACKGROUND: The efficiency of patient safety interventions is not well studied, especially laboratory monitoring for drug therapy. More than one-third of preventable adverse drug events are associated with inadequate monitoring. Current knowledge of decreasing adverse drug events through expanded monitoring programs is lacking. DESIGN: The authors focused on a laboratory monitoring program (above usual practice) of renin-angiotensin system (RAS) agents to prevent adverse events of hyperkalemia and acute renal failure. They used a probabilistic decision model to estimate cost savings and cost effectiveness (at $30,000 and $10,000 per quality-adjusted life-year (QALY)). Costs included the monitoring program, and offsets from reduced care in 3 populations (overall, chronic kidney disease [CKD], and diabetes). MAIN RESULTS: Adverse events were most common in those with CKD. Intervening on all new users or the subset with diabetes was almost never expected to be cost saving (probability <1%). But a monitoring program restricted to patients with CKD was expected to be cost saving (probability = 95%). A strategy that intervened on all patients, or those with diabetes, was never cost effective, (probability <1%). But intervening on patients with CKD was estimated to be cost effective (at either cost-effectiveness threshold) at least 95% of the time in the base case. CONCLUSIONS: The authors' findings illustrate that for laboratory monitoring to be cost effective, the patient population must be at high enough risk of adverse events. Further inquiry into the willingness to pay for patient safety interventions is needed.

Active influenza vaccine safety surveillance: potential within a healthcare claims environment.

BACKGROUND: Rapid safety assessment of novel vaccines, especially those targeted against pandemic influenza, is a public health priority. OBJECTIVES: Assess the feasibility of using healthcare claims data to rapidly detect influenza vaccine adverse events using sequential analytic methods. RESEARCH DESIGN: Retrospective pilot study simulating prospective surveillance using 6 cumulative monthly administrative claims data extracts. The first included encounters occurring in October; each subsequent extract included an additional month of encounters. Ten adverse events were evaluated, comparing postvaccination rates during the 2006-2007 influenza season to those expected based on rates observed in the prior season. SUBJECTS: Members of a large, multistate health insurer who had a claim for influenza vaccination during the 2005-2006 or 2006-2007 influenza seasons. MEASURES: The completeness of monthly claims extracts. RESULTS: Most vaccinations and outcomes were identified early in the 2006-2007 season; about 50% of vaccinations and short latency events were identified in the second monthly data extract, which would typically become available by mid-December, and 80% of vaccinations and events were identified in the third extract. With respect to overall claims lag, approximately 90% of vaccinations and events were identified within 1 to 2 months after vaccination, regardless of vaccination month. CONCLUSIONS: This study suggests that administrative claims data might contribute to same season influenza vaccine safety surveillance in large, defined populations, especially during a threat of pandemic influenza. Based on our previous work, we believe this method could be applied to multiple health plans' data to monitor a large portion of the US population.

Statin use and risk of colorectal cancer in a cohort of middle-aged men in the US: a prospective cohort study.

BACKGROUND: Numerous modifiable factors have been associated with a reduced risk of colorectal cancer, including the chronic use of NSAIDs. Thus, it is biologically plausible that HMG-CoA reductase inhibitors (statins), therapeutic agents that also possess anti-inflammatory effects, are also associated with a lowered risk of colorectal cancer. OBJECTIVE: To examine the association between statin use and the risk of colorectal cancer in a large cohort of middle-aged men enrolled in a prepaid, integrated health maintenance organization. METHODS: We conducted a prospective cohort study of 69 115 Northern and Southern California Kaiser Permanente (KP) members aged 45-69 years who enrolled in the California Men's Health Study in 2002-3. Colorectal cancer cases were identified by linkage to the KP California Cancer Registries. Statin exposure, estimated from automated KP outpatient pharmacy records (available since 1991 in Southern California and 1994 in Northern California), was treated as time-varying. Cox proportional hazards regression analyses were used to estimate hazard ratios and 95% confidence intervals (CIs), while controlling for potential confounders. RESULTS: During a maximum of 3.5 years of follow-up, 171 colorectal cancer cases were identified. Compared with nonuse, the adjusted hazard ratio for ever use of statins was 0.89 (95% CI 0.61, 1.30). The hazard ratio for statin use of >or=5 years was 0.83 (95% CI 0.43, 1.63). The results did not differ markedly by type or severity of disease. There was also no evidence of effect modification by regular NSAID use. However, the stratified analyses were limited by small numbers. CONCLUSION: These findings provide little support for an association between the use of statins and the risk of colorectal cancer in men. There was some suggestion of a modest inverse association between statin use for >or=5 years and risk of colorectal cancer; however, the possibility that this observation may be related to regular NSAID use cannot be ruled out.

Medication adherence and use of generic drug therapies.

OBJECTIVE: To assess whether lower copayments charged for generic drugs explain the improved drug adherence associated with use of generics. METHODS: We analyzed 2001-2004 healthcare claims data from 45 large employers. Study subjects were age > or = 18 years, had 1 or more of 5 study conditions (hypercholesterolemia, hypertension, hypothyroidism, seizure disorders, type 2 diabetes), and new use of generic-only or brand-only drug therapy for that condition. We measured adherence as the medication possession ratio (MPR), and adequate adherence as MPR > or = 80%. Logistic regressions were conducted to assess adequate adherence, adjusting for copayments. RESULTS: We identified 327,629 new users of drug therapy. The proportion starting generic therapies ranged from 9% (hypothyroidism) to 45% (hypertension). After 1 year, 66.2% of individuals with hypothyroidism achieved an MPR > or = 80% compared with 53.4% with hypertension, 53.2% with hypercholesterolemia, 52.0% with diabetes, and 42.2% with seizure disorders. Generics were associated with greater adherence than brands in patients with hypercholesterolemia or diabetes (P <.05). Lower adherence was seen in patients with hypertension or hypothyroidism (P <.05). There was no difference in seizure disorders. The likelihood of achieving an MPR =80% with $0 copayments compared with $1 to $9 ranged from an adjusted odds ratio (AOR) of 1.32 for seizure disorders (95% confidence interval [CI] = 1.41, 1.43) to an AOR of 1.45 for hypothyroidism (95% CI = 1.43, 1.48). CONCLUSION: Generic prescribing was associated with modestly improved adherence in 2 of 5 study conditions. Copayments of $0 were associated with improved adherence across all conditions.

Early adverse drug event signal detection within population-based health networks using sequential methods: key methodologic considerations.

PURPOSE: Active surveillance of population-based health networks may improve the timeliness of detection of adverse events (AEs). Our objective was to expand our previous signal detection work by investigating the effect on signal detection of alternative study specifications. METHODS: We compared the signal detection performance under various study specifications using historical data from nine health plans involved in the HMO Research Network's Center for Education and Research on Therapeutics (CERT). Five drug-event pairs representing generally accepted associations with an AE and two pairs representing "negative controls" were analyzed. Alternative study specifications related to the definition of incident users and incident AEs were assessed and compared to our previous findings. RESULTS: Relaxing the incident AE exclusion criteria by (1) including members with prior outpatient diagnoses of interest and (2) halving (to 90 days) the time window specified to define incident exposure and diagnoses increased the number of members under surveillance and as a consequence increased the number of exposed days and diagnoses by about 10-20%. The alternative specifications tend to result in earlier signal detection by 10-16 months, a likely consequence of more exposures and events entering the analysis. CONCLUSIONS: This paper provides additional preliminary information related to conducting prospective safety monitoring using health plan data and sequential analytic methods. Our findings support continued investigation of using health plan data and sequential analytic methods as a potentially important contribution to active drug safety surveillance.

Hormone therapy initiation after the Women's Health Initiative.

OBJECTIVES: To describe hormone therapy (HT) initiation after the 2002 publication of the Women's Health Initiative. DESIGN: Observational cohort (1999-2003) of women ages 40 to 79 years, five health plans, used HT in July 2002 and subsequently discontinued or never used before August 2002. RESULTS: Of discontinuers, 15.8% (3,203 of 20,205) reinitiated HT. Reinitiation was higher among estrogen users (23.8%) versus estrogen with progestin users (11.3%), and lower among those with diabetes (relative risk [RR]=0.68, 95% CI: 0.61-0.76), cardiovascular disease (RR=0.87, 95% CI: 0.83-0.92), and hyperlipidemia (RR=0.83, 95% CI: 0.79-0.88). Only 2.3% (2,072 of 90,261) of never users initiated (August 2002 to December 2003). First-time initiation was associated with cardiovascular disease (RR=1.17, 95% CI: 1.10-1.25) and hyperlipidemia (RR=1.24, 95% CI: 1.17-1.33) and was less common among those with diabetes (RR=0.70, 95% CI: 0.63-0.79). CONCLUSIONS: After the Women's Health Initiative, a minority of women reinitiated or became first-time initiators of HT. Women with cardiovascular disease, diabetes, and hyperlipidemia were less likely to reinitiate; women with cardiovascular disease and hyperlipidemia were more likely to be first-time initiators.

Statin use and risk of prostate cancer in the California Men's Health Study cohort.

Statins have known anticarcinogenic effects, however, evidence for long-term statin use as effective chemoprevention for prostate cancer is inconsistent. We examined the association between statin use and risk of prostate cancer among 69,047 eligible participants in the California Men's Health Study, a prospective cohort of Northern and Southern California Kaiser Permanente (KP) members, ages 45 to 69 years, initiated in 2002. Prostate cancer cases were identified by linkage to the KP California Cancer Registries. Statin exposure, estimated from automated KP outpatient pharmacy records (available since 1991 in Southern California and since 1994 in Northern California), was treated as time-varying and defined as the cumulative days dispensed of any statin from the first dispensing until a prostate cancer diagnosis, radical prostatectomy, termination of membership, or end of study (December 31, 2004). Cox proportional hazards models with age as the time scale were used to estimate rate ratios, while controlling for confounding variables. During follow-up, 888 prostate cancer cases, including 131 advanced cases, were identified. There was no association between ever statin use or <5 years use and prostate cancer. Conversely, >or=5 years use was associated with a 28% lower risk for prostate cancer compared with nonuse (adjusted rate ratio, 0.72; 95% confidence interval, 0.53-0.99). This association did not differ markedly for advanced disease. However, the association did seem to be restricted to those who regularly take nonsteroidal anti-inflammatory drugs. Our findings suggest that long-term statin use might be associated with a reduced risk of prostate cancer but perhaps only among regular nonsteroidal anti-inflammatory drug users.

Early detection of adverse drug events within population-based health networks: application of sequential testing methods.

PURPOSE: Active surveillance of population-based health networks may improve the timeliness of detection of adverse drug events (ADEs). Active monitoring requires sequential analysis methods. Our objectives were to (1) evaluate the utility of automated healthcare claims data for near real-time drug adverse event surveillance and (2) identify key methodological issues related to the use of healthcare claims data for real-time drug safety surveillance. METHODS: We assessed the ability to detect ADEs using historical data from nine health plans involved in the HMO Research Network's Center for Education and Research on Therapeutics (CERT). Analyses were performed using a maximized sequential probability ratio test (maxSPRT). Five drug-event pairs representing known associations with an ADE and two pairs representing 'negative controls' were analyzed. RESULTS: Statistically significant (p < 0.05) signals of excess risk were found in four of the five drug-event pairs representing known associations; no signals were found for the negative controls. Signals were detected between 13 and 39 months after the start of surveillance. There was substantial variation in the number of exposed and expected events at signal detection. CONCLUSIONS: Prospective, periodic evaluation of routinely collected data can provide population-based estimates of medication-related adverse event rates to support routine, timely post-marketing surveillance for selected ADEs.

Validity of gout diagnoses in administrative data.

OBJECTIVE: To determine the utility of using administrative data for epidemiologic studies of gout by examining the validity of gout diagnoses in claims data. METHODS: From a population of approximately 800,000 members from 4 managed care plans, we identified patients who had at least 2 ambulatory claims for a diagnosis of gout between January 1, 1999 and December 31, 2003. From this group, a random sample of 200 patients was chosen for medical record review. Trained medical record reviewers abstracted gout-related clinical, laboratory, and radiologic data from the medical records. Two rheumatologists independently evaluated the abstracted information and assessed whether the gout diagnosis was probable/definite or unlikely/insufficient information. Discordant physician ratings were adjudicated by consensus. Based on record reviews, patients were also classified according to the American College of Rheumatology (ACR), Rome, and New York gout criteria and these results were compared with the physician global assessments. RESULTS: There were 121 patients rated as having probable/definite gout by physician consensus, leading to a positive predictive value of >or=2 coded diagnoses of gout of 61% (95% confidence interval 53-67). There was low concordance between physician assessments and established gout criteria including ACR, Rome, and New York criteria (kappa = 0.17, 0.16, and 0.20, respectively). CONCLUSION: Use of administrative data alone in epidemiologic and health services research on gout may lead to misclassification. Medical record reviews for validation of claims data may provide an inadequate gold standard to confirm gout diagnoses.

Potential drug-drug interactions in the outpatient setting.

BACKGROUND: Although medication safety research has tended to focus on inpatients, the safety of drug use among outpatients is also a concern. OBJECTIVE: We estimate the frequency of potentially interacting concomitant medication dispensing among outpatients. RESEARCH DESIGN: We report the number and percent of patients annually dispensed an object drug of interest (ie, warfarin, digoxin, cyclosporine, or lovastatin/simvastatin) with a potentially interacting drug among a random sample of insured adults receiving care from 10 integrated delivery systems. We use 2 definitions of concomitant dispensing: medications dispensed: 1) during the time period for which the patient had the other medication available ('days supply') and 2) on the same day. We also estimate the number of insured U.S. population codispensed these medication pairs. RESULTS: Among patients dispensed a drug of interest, between 17.8% (95% confidence interval [CI]=17.1-18.6%) and 28.0% (95% CI=24.0-32.1%) were concomitantly dispensed a potentially interacting drug using the "days supply" definition, and between 7.1% (95% CI=6.6-7.7%) and 17.7% (95% CI=14.4-21.1%) using the "same day" definition. Extrapolating to the insured U.S. population, between 1.29 (95% CI=1.25-1.33; same day) and 2.67 (95% CI=2.62-2.72; days supply) million insured adults are dispensed 1 of these 4 potentially interacting pairs. CONCLUSIONS: We found evidence of potentially interacting concomitant medication dispensing among outpatients. An opportunity exists to better understand how such dispensing translates into adverse events and ultimately to improved medication safety.

Use of prescription medications with a potential for fetal harm among pregnant women.

PURPOSE: To estimate the prevalence of use of prescription drugs with a potential for fetal harm among pregnant women in the United States. METHODS: A retrospective study was conducted using the automated databases of eight health maintenance organizations involved in the HMO Research Network Center for Education and Research on Therapeutics (CERT). Women who delivered an infant from January 1996 to December 2000 were identified. The frequency of use of prescription drugs with a potential for fetal harm was based upon the expert review of a clinical teratologist and the U.S. Food and Drug Administration (FDA) risk classification system, assuming a gestational duration of 270 days. RESULTS: Among the 114 165 women with no documentation of a diagnosis suggesting potential pre-term birth or dispensing of ovulation stimulants in the 270 days before delivery, 1305 (1.1%) received a teratogenic drug during the 270 days before delivery, based upon the expert review of a clinical teratologist. A larger proportion of women received U.S. FDA category D or X drugs (5.8%; N = 6600). However, the general patterns of use were similar, with higher use in early pregnancy compared to later trimesters. The proportion of women dispensed a teratogen during pregnancy was substantially higher among women who received a teratogen in the 90 days before pregnancy compared to women who did not (adjusted RR = 38.9, 95%CI, 33.5, 45.3). CONCLUSIONS: Our results suggest that further efforts directed at physicians to counsel women or at the women themselves about the potential risks of particular medications appear warranted.

Frequency of serum creatinine monitoring during allopurinol therapy in ambulatory patients.

BACKGROUND: Allopurinol dosage reduction is recommended in patients with renal dysfunction because drug toxicity risk is increased. Little information is available about serum creatinine (SCr) monitoring in ambulatory patients taking allopurinol. OBJECTIVE: To evaluate SCr monitoring among patients prescribed allopurinol, identify associated factors, and evaluate administrative data in assessing monitoring. METHODS: Information for this retrospective cohort study was drawn from a dataset of 2 020 037 individuals; approximately 200 000 members from each of 10 organizations. Study patients had received at least one year of ongoing allopurinol prescription dispensings. Patient variables analyzed included age, gender, chronic diseases, outpatient visits, hospitalizations, gout diagnosis, and SCr monitoring. A random sample of medical records was reviewed to assess the accuracy of the automated data. Statistical analysis included descriptive and logistic regression techniques. RESULTS: Overall, 1139 (26%) of 4357 patients did not have SCr monitoring. For individuals without recent hospitalization, factors protective against lack of monitoring were increasing age (OR 0.77 per 10 y; 95% CI 0.74 to 0.79), more chronic diseases (OR 0.81; 95% CI 0.78 to 0.83), more outpatient visits (OR 0.87 per 5 visits; 95% CI 0.83 to 0.91), and gout diagnosis (OR 0.74; 95% CI 0.65 to 0.85). The sensitivity and specificity of administrative data compared with medical records for SCr monitoring were 92% and 65%, respectively. CONCLUSIONS: More than one-fourth of patients dispensed allopurinol did not have SCr monitoring during one year of therapy. Lack of monitoring and lack of subsequent possible dosage adjustment put patients at increased risk of allopurinol toxicity.