Brand and generic use of inhalation medication and frequency of switching in children and adults: A population-based cohort study.

BACKGROUND: The expiration of patents of brand inhalation medications and the ongoing pressure on healthcare budgets resulted in a growing market for generics. AIM: To study the use of brand and generic inhalation medication and the frequency of switching between brand and generic and between devices. In addition, we investigated whether switching affected adherence. METHODS: From dispensing data from the Dutch PHARMO Database Network a cohort aged ≥ 5 years, using ≥ 1 year of inhalation medication between 2003 and 2012 was selected. Switching was defined as changing from brand to generic or vice versa. In addition, we studied change in aerosol delivery device type (e.g., DPI, pMDI, and nebulizers). Adherence was calculated using the medication possession ratio (MPR). RESULTS: The total cohort comprised 70,053 patients with 1,604,488 dispensations. Per calendar year, 5% switched between brand and generic inhalation medication and 5% switched between devices. Median MPRs over the first 12 months ranged between 33 and 55%. Median MPR over the total period was lower after switch from brand to generic and vice versa for formoterol (44.5 vs. 42.1 and 63.5 vs. 53.8) and beclomethasone (93.8 vs. 59.8 and 81.3 vs. 55.9). CONCLUSION: Per year, switching between brand and generic inhalation medication was limited to 5% of the patients, switching between device types was observed in 5% as well. Adherence to both generic and brand inhalation medication was low. Effect of switching on adherence was contradictory; depending on time period, medication and type, and direction of switching. Further research on reasons for switching and potential impact on clinical outcomes is warranted.

Quick assessment of drug-related admissions over time (QUADRAT study).

PURPOSE: To develop a computerized prescreening procedure for the identification of possible/probably Hospital Admissions potential Related to Medications (HARMs). METHOD: Pairs of drugs and reasons for hospitalization (generated automatically from the PHARMO record linkage database by using two data mining techniques) were assessed manually to determine whether they represented pharmacologically plausible adverse drug events (PP-ADEs). Two crude samples of these PP-ADEs (from 2005 and 2008) were examined manually to establish causality and preventability on the basis of hospital discharge letters plus medication dispensing data. The results were used to calculate the positive predictive value (PPV) of the crude causality PP-ADEs, the net percentage of possible/probably HARMs, and their potential preventability. RESULTS: Data mining by Gamma Poisson Shrinkage and trend analysis produced 1330 and 2941 significant drug-event pairs, respectively. After manual assessment, 307 different PP-ADEs remained. The annual prevalence of these PP-ADEs was stable at approximately 8% throughout 2000-2009. Manual assessment of two samples of crude PP-ADEs showed that their causality PPV was 53.7% (95%CI: 52.7%-54.7%) in 2005 and 47.9% (95%CI: 46.9%-49.0%) in 2008. The net contribution of possible/probably HARMs to all acute admissions was 4.6% (95%CI: 4.5%-4.8%) in 2005 and 3.9% (95%CI: 3.8%-4.0%) in 2008. The potential preventability of all possible/probably HARMs in the two samples was 19.3% (95%CI: 18.5-20.1). CONCLUSION: Automated pre-selection of PP-ADEs is an efficient way to monitor crude trends. Further validation and manual assessment of the automatically selected hospitalizations is necessary to get a more detailed and precise picture.

Antidepressant use and the risk of suicide: a population-based cohort study.

BACKGROUND: The existing literature provides contradictory evidence on antidepressant use and risk of suicide. Some studies have shown that the use of Selective Serotonin Reuptake Inhibitors (SSRIs) is associated with an increased risk of suicide, especially during the first months of treatment, whereas other studies did not confirm this association. For this reason, our objective was to investigate the association between antidepressant use and risk of suicide in incident antidepressant users in relation to time since starting therapy. METHODS: We conducted a population-based cohort study within the Dutch Integrated Primary Care Information (IPCI) database, in incident users of antidepressant therapy between 1994 and 2012 (n=27,712). Cox proportional hazard models were used to study the association between current use of SSRIs, tricyclic antidepressants (TCA) and other antidepressants and risk of suicide or attempted suicide. RESULTS: During follow-up, a total of 280 incident antidepressant users attempted or committed suicide. Current use of SSRIs (hazard ratio (HR): 0.78, 95% CI: 0.57-1.07), TCAs (HR: 0.82, 95% CI: 0.48-1.42) or other antidepressants (HR: 0.75, 95% CI: 0.47-1.18) was not statistically significantly associated with suicide compared to past use of any of the antidepressants. LIMITATIONS: Although a large healthcare database was used, the number of reported cases of suicide (attempt) was low. CONCLUSIONS: This study did not indicate an increase in risk of suicide after starting treatment with SSRIs, TCAs or other antidepressants compared with past antidepressant use.

Risk of hyponatremia with diuretics: chlorthalidone versus hydrochlorothiazide.

BACKGROUND: Chlorthalidone and hydrochlorothiazide are often considered as interchangeable. However, greater (nighttime) blood pressure reduction, and alleged pleiotropic effects have renewed the interest in chlorthalidone. A recent study showed an increased risk of adverse events with chlorthalidone, including hyponatremia. METHODS: To investigate differences in risk of hyponatremia between chlorthalidone and hydrochlorothiazide, adjusted for daily dose, we conducted a population-based case-control study within the Dutch IPCI (Integrated Primary Care Information) database. The study population included all subjects ≥18 years without diabetes mellitus, heart failure, liver failure, and malignancy, who were registered in the IPCI database from 1996 to 2011. Cases were subjects with a serum sodium <130 millimoles per liter or hospitalization due to hyponatremia. Controls were matched on practice, age within 5 years, sex, and date of onset. RESULTS: A total of 1033 cases of hyponatremia were identified. Hyponatremia was more common with chlorthalidone than with hydrochlorothiazide at equal dose per day: adjusted odds ratio was 2.09 (95% confidence interval [CI], 1.13-3.88) for 12.5 milligrams per day and 1.72 (95% CI, 1.15-2.57) for 25 milligrams per day. Risks were not significantly increased with chlorthalidone compared with twice the dose per day of hydrochlorothiazide. CONCLUSIONS: This is the first study that shows an increased risk of hyponatremia with chlorthalidone relative to hydrochlorothiazide at equal milligram-to-milligram dose per day. The need for a lower dose of chlorthalidone than hydrochlorothiazide to achieve similar blood pressure reduction likely compensates for the increased risk of hyponatremia at equal dose.

Adherence to renal function monitoring guidelines in patients starting antihypertensive therapy with diuretics and RAAS inhibitors: a retrospective cohort study.

BACKGROUND: Acute kidney injury (AKI) might complicate antihypertensive therapy. In The Netherlands, general practitioner clinical practice guidelines provide clear recommendations on monitoring of renal function to minimize this risk. Our objective was to investigate how day-to-day clinical practice corresponds to the guidelines. METHODS: We conducted a retrospective cohort study in a dynamic population, using data on >9,000 adults that was retrieved from the Integrated Primary Care Information database. We investigated whether serum creatinine (SCR) was measured within 30 and 365 days after the start of (combined) use of a diuretic, an angiotensin-converting enzyme inhibitor, and/or angiotensin receptor blocker. We also investigated the association between calendar year, sex, type of therapy, risk factors for AKI and practice and SCR measurement. RESULTS: Of 6,593 subjects who met the study criteria for single drug therapy, SCR was measured in 1,233 subjects within 30 days and in 3,896 subjects within 365 days. For combined drug therapy recipients (n = 2,497), these were 545 and 1,687, respectively. Associated cumulative probabilities were 19 % and 66 % with single drug therapy, and 22 % and 74 % with combined drug therapy. Significant differences were observed between practices. SCR measurement was associated with other characteristics, except for sex. Within 365 days, SCR increased >30 % of baseline in 103 subjects (1.6 %) after the start of single drug therapy, and in 85 (3.4 %) subjects who initiated combined drug therapy. In the majority (>70 %) of these subjects, this did not result in subsequent monitoring or adjustment of antihypertensive treatment. CONCLUSIONS: Results from this study suggest that, on average, renal function is not monitored as strictly as recommended by relevant clinical practice guidelines.

A population-based study on the prevalence and incidence of chronic kidney disease in the Netherlands.

PURPOSE: Because most population-based studies on the epidemiology of chronic kidney disease (CKD) are cross-sectional, there is, except for end-stage renal disease, hardly any information on incidence rates. METHODS: We conducted a retrospective cohort study in a dynamic population, using data of 784,563 adult participants retrieved from the Integrated Primary Care Information database, a primary care database containing the complete electronic longitudinal medical records. CKD (both incidence and prevalence) was based on (1) an increased urine albumin-to-creatinine ratio, (2) a decreased estimated glomerular filtration rate, or (3) explicit statement in the medical record. Results were stratified by age according to the WHO standard population, sex, and diabetes mellitus. RESULTS: Based on a single measurement only, the incidence rate of CKD in adults was 1,213 per 100,000 person-years, and 6.7 percent of the adult population had a prevalent diagnosis of CKD. The incidence rate increased by age and was the highest in participants with diabetes with an incidence of 25,000 per 100,000 person-years, affecting over 75 percent of participants with diabetes. CONCLUSIONS: This is the first study to report the incidence rates of all stages of CKD for the entire adult population, stratified by sex, 5-year age groups, and diabetes. Our data demonstrate that the incidence of CKD increases with age and is the highest in participants with diabetes mellitus.

Improving sensitivity of machine learning methods for automated case identification from free-text electronic medical records.

BACKGROUND: Distinguishing cases from non-cases in free-text electronic medical records is an important initial step in observational epidemiological studies, but manual record validation is time-consuming and cumbersome. We compared different approaches to develop an automatic case identification system with high sensitivity to assist manual annotators. METHODS: We used four different machine-learning algorithms to build case identification systems for two data sets, one comprising hepatobiliary disease patients, the other acute renal failure patients. To improve the sensitivity of the systems, we varied the imbalance ratio between positive cases and negative cases using under- and over-sampling techniques, and applied cost-sensitive learning with various misclassification costs. RESULTS: For the hepatobiliary data set, we obtained a high sensitivity of 0.95 (on a par with manual annotators, as compared to 0.91 for a baseline classifier) with specificity 0.56. For the acute renal failure data set, sensitivity increased from 0.69 to 0.89, with specificity 0.59. Performance differences between the various machine-learning algorithms were not large. Classifiers performed best when trained on data sets with imbalance ratio below 10. CONCLUSIONS: We were able to achieve high sensitivity with moderate specificity for automatic case identification on two data sets of electronic medical records. Such a high-sensitive case identification system can be used as a pre-filter to significantly reduce the burden of manual record validation.