Search

Author(s): Jennings, Lee A; Hollands, Simon; Keeler, Emmett; Wenger, Neil S; Reuben, David B | Abstract: Background/objectivesAlthough nurse practitioner dementia care co-management has been shown to reduce total cost of care for fee-for-service (FFS) Medicare beneficiaries, the reasons for cost savings are unknown. To further understand the impact of dementia co-management on costs, we examined acute care utilization, long-term care admissions, and hospice use of program enrollees as compared with persons with dementia not in the program using FFS and managed Medicare claims data.DesignQuasi-experimental controlled before-and-after comparison.SettingUrban academic medical center.ParticipantsA total of 856 University of California, Los Angeles (UCLA) Alzheimer's and Dementia Care program patients were enrolled between July 1, 2012, and December 31, 2015, and 3,139 similar UCLA patients with dementia not in the program. Comparison patients were identified as having dementia using International Classification of Diseases-9 codes and natural language processing of clinical notes. Coarsened exact matching was used to reduce covariate imbalance between intervention and comparison patients.InterventionDementia co-management model using nurse practitioners partnered with primary care providers and community organizations.MeasurementsAverage difference-in-differences per quarter over the 2.5-year intervention period for all-cause hospitalization, emergency department (ED) visits, intensive care unit (ICU) stays, and number of inpatient hospitalization days; admissions to long-term care facilities; and hospice use in the last 6 months of life.ResultsIntervention patients had fewer ED visits (odds ratio [OR] = .80; 95% confidence interval [CI] = .66-.97) and shorter hospital length of stay (incident rate ratio = .74; 95% CI = .55-.99). There were no significant differences between groups for hospitalizations or ICU stays. Program participants were less likely to be admitted to a long-term care facility (hazard ratio = .65; 95% CI = .47-.89) and more likely to receive hospice services in the last 6 months of life (adjusted OR = 1.64; 95% CI = 1.13-2.37).ConclusionComprehensive nurse practitioner dementia care co-management reduced ED visits, shortened hospital length of stay, increased hospice use, and delayed admission to long-term care.

Purpose: To determine whether the use of worksite health and wellness clinics reduced hemoglobin A1c (HbA1c) for prediabetic and diabetic employees. Design: Hemoglobin A1c values were compared between clinic users and matched non-users. Setting: The Wonderful Company’s (TWC’s) agricultural and packaging sites in Central California. Sample: TWC’s 2016 to 2017 employees who used clinics (n = 445, defined below) and clinic non-users (n = 217) who had HbA1c screening and worked at TWC for 3 or more months each year. Intervention: A unique worksite health and wellness clinic that offers multidisciplinary primary medical care in synergy with TWC’s overall wellness programs. Measures: Exposure was clinic use, defined by frequency and patterns of visits. Hemoglobin A1c was the outcome of interest. Analysis: Users and non-users were propensity score matched using the 2016 employee data including HbA1c, and then 2017 HbA1c values were compared between the 2 groups. Results: The 2017 HbA1c of diabetic employees was lower among clinic users compared to non-users (7.42 vs 8.53, P < .001). Differences in HbA1c among prediabetics or diabetics and prediabetics aggregated were not statistically significant, despite TWC’s population-level data showing a reduction in prediabetes prevalence. The clinic impact results were robust to multivariate analyses and an alternative definition of utilization. Conclusion: The implementation of TWC’s Health and Wellness clinics was associated with reductions in HbA1c among diabetics, but further research is needed on prediabetics.

Pharmaceutical promotion can lead to market size expansion, which is beneficial if previously untreated patients access treatment but deleterious if it leads to overuse, an area of concern for second generation antipsychotics (SGA). We contribute to a growing body of work suggesting that networks of social and professional relationships shape prescribing behavior. We examined 88,439 Medicare Part D prescribing physicians, finding that promotion is associated with SGA market size expansion (elasticity: 0.062) and that network-level promotional activity is associated with network members' branded product prescribing. Research on the effects of promotion should account for its effects in prescribers' networks.

Theory suggests that disability discrimination protections may adversely affect the hiring of individuals with disabilities by making them more expensive. Using SIPP data, we explore how the ADA Amendments Act (ADAAA), which expanded disability discrimination protections, affected the relative hiring rate of individuals with disabilities. We employ new categorizations of disability type: salient physical conditions, non-salient physical conditions, mental retardation and developmental disability, and other mental conditions. We find that the ADAAA had no effect other than increasing hiring for those with non-salient physical conditions. These results suggest that condition saliency may mediate the effects of discrimination protections on hiring.

Objective To report on patterns of optometry prescribing and adherence to prescribing regulations at a tertiary care ophthalmology centre. Design Prospective cohort study. Participants All new referrals from optometrists and other health care professionals to the emergency eye care service (n = 1965) between July 2011 and June 2012, as well as optometry referrals to 2 subspecialty services (glaucoma [n = 71] and cornea [n = 212]). Methods In our primary analysis, the frequency of prescribing and classes of medications prescribed were reported and compared amongst various referral sources to the emergency eye care service. As a secondary analysis, we reported frequency of prescribing in optometry referrals to 2 subspecialty clinics. Adherence to prescribing guidelines was reported for all optometry referrals. Results Of 296 referrals from optometrists to the emergency eye care service, 20 (6.8%) had received a prescription medication; this was significantly less compared to emergency and family physicians ( p Conclusions After the introduction of prescribing privileges, optometrists prescribe a variety of medications in their practices. Current prescribing guidelines are not always followed in practice, suggesting that continued study and collaboration is necessary to create an optimal model of interdisciplinary care that provides access to the highest quality of care.

Objective To evaluate excess mortality over time, comparing rheumatoid arthritis (RA) patients with the general population. Methods We computed all-cause mortality rates among Ontario residents age ≥15 years with RA versus without RA from 1996 to 2009. Age- and sex-standardized mortality rates were expressed as the number of deaths per 1,000 population. Excess mortality rates were calculated as the difference between death rates among RA patients and those in the general population. We estimated standardized mortality ratios (SMRs) and mortality rate ratios (MRRs) to assess relative excess mortality over time. Results From 1996 to 2009, SMRs in RA ranged from 13.0 (95% confidence interval [95% CI] 12.2, 13.9) to 9.2 deaths per 1,000 RA patients (95% CI 8.4, 10.0); and for those without RA from 8.7 (95% CI 8.6, 8.7) to 6.0 deaths (95% CI 5.9, 6.0) per 1,000 general population. Over the study period, the excess mortality rate among RA patients was approximately 3 excess deaths per 1,000 population. Relative reductions in standardized mortality rates occurred over time for those with and without RA (−21.4% versus −13.4%). The SMRs for RA patients in 1996–1997, 2000–2001, 2004–2005, and 2008–2009 were 1.51 (95% CI 1.43, 1.59), 1.50 (95% CI 1.43, 1.57), 1.43 (95% CI 1.37, 1.50), and 1.41 (95% CI 1.35, 1.47), respectively. We did not find a significant change in the MRR by calendar time. Conclusion Mortality for RA patients has decreased over time but remains elevated compared to the general population, with 40–50% more deaths among RA patients. The relative excess mortality over time (mortality gap) remains unchanged in our sample.

Background: Trimethoprim–sulfamethoxazole increases the risk of hyperkalemia when used with spironolactone. We examined whether this drug combination is associated with an increased risk of sudden death, a consequence of severe hyperkalemia. Methods: We conducted a population-based nested case–control study involving Ontario residents aged 66 years or older who received spironolactone between Apr. 1, 1994, and Dec. 31, 2011. Within this group, we identified cases as patients who died of sudden death within 14 days after receiving a prescription for trimethoprim–sulfamethoxazole or one of the other study antibiotics (amoxicillin, ciprofloxacin, norfloxacin or nitrofurantoin). For each case, we identified up to 4 controls matched by age and sex. We determined the odds ratio (OR) for the association between sudden death and exposure to each antibiotic relative to amoxicillin, adjusted for predictors of sudden death using a disease risk index. Results: Of the 11 968 patients who died of sudden death while receiving spironolactone, we identified 328 whose death occurred within 14 days after antibiotic exposure. Compared with amoxicillin, trimethoprim–sulfamethoxazole was associated with a more than twofold increase in the risk of sudden death (adjusted OR 2.46, 95% confidence interval [CI] 1.55–3.90). Ciprofloxacin (adjusted OR 1.55, 95% CI 1.02–2.38) and nitrofurantoin (adjusted OR 1.70, 95% CI 1.03–2.79) were also associated with an increased risk of sudden death, although the risk with nitrofurantoin was not apparent in a sensitivity analysis. Interpretation: The antibiotic trimethoprim–sulfamethoxazole was associated with an increased risk of sudden death among older patients taking spironolactone. When clinically appropriate, alternative antibiotics should be considered in these patients.

Objective Therapeutic intraarticular injections are used in the management of hip osteoarthritis (OA). Some studies suggest that their use increases the risk of infection and subsequent revision surgery after primary total hip arthroplasty (THA), while others do not. We undertook this study to clarify the relationship between prior intraarticular injection and the risk of complication in a subsequent primary THA. Methods In a cohort of patients with hip OA who underwent a primary elective THA between 2002 and 2009, we identified those who received ≥1 intraarticular injection performed by a radiologist in the 5 years preceding their THA. Multivariable Cox proportional hazards models were used to determine the relationship between receipt of a presurgical injection (no injection, 1–5 years prior to THA, or

This report estimates the potential impacts of a California Workers' Compensation formulary in terms of changes in prescription drug use and spending and discusses the potential broader impacts of these changes on the California economy. This report is the second of three related reports.

Objective To determine the relationship between surgeon experience with, and complications following, total joint arthroplasty (TJA) in patients with rheumatoid arthritis (RA). Methods Using administrative data, we assembled a cohort of patients with RA who had undergone at least 1 elective primary hip or knee replacement procedure between 2002 and 2009. Cox proportional hazards, censored on death and accounting for clustering of patients within surgeons, were used to determine the relationship between overall and “RA-specific” surgeon TJA volume and the occurrence of a composite “complication” outcome (revision, infection, dislocation, or periprosthetic fracture within 2 years of the initial TJA), controlling for potential confounders (patient age, sex, comorbidity, and disease severity). Results We identified 4,762 patients with RA who were eligible for TJAs (1,515 total hip arthroplasties and 3,247 total knee arthroplasties). Among these patients, 152 (3.2%) experienced a surgical complication within 2 years of the procedure. After controlling for patient and hospital factors, greater surgeon TJA volume in patients with RA (RA TJA), but not overall TJA volume (all TJA), was associated with a reduced risk of complications (for surgeon RA TJA volume per 10 cases, adjusted hazard ratio [HR] 0.81, 95% confidence interval [95% CI] 0.71–0.93, P = 0.002; for surgeon all TJA volume, adjusted HR 0.98, 95% CI 0.97–1.00, P = 0.09). Conclusion In a cohort of patients with RA who underwent hip or knee TJA, increased surgeon experience performing TJA in patients with RA, irrespective of their overall TJA experience and hospital factors, was associated with a decreased risk of surgical complications. These findings have potential implications for surgeon training and the referral practices of rheumatologists.

Objective Most of the evidence regarding complications following total hip arthroplasty (THA) and total knee arthroplasty (TKA) are based on patients with osteoarthritis (OA); less is known about outcomes in rheumatoid arthritis (RA). Using a validated algorithm for identifying patients with RA, we undertook this study to compare the rates of complications among THA and TKA recipients between those with RA and those without RA. Methods In patients who underwent a first primary elective THA or TKA between 2002 and 2009, those with RA were identified using a validated algorithm: a hospitalization with a diagnosis code for RA or 3 physician billing claims with a diagnosis code for RA, with at least 1 claim by a specialist (rheumatologist, orthopedic surgeon, or internist) in a 2-year period. Recipients with diagnostic codes suggesting an inflammatory arthritis, but not meeting RA criteria, were classified as having inflammatory arthritis. All remaining patients were deemed to have OA. Cox proportional hazards models, censored on death, were used to determine the relationship between the type of arthritis and the occurrence of specific complications, adjusting for potential confounders (age, sex, comorbidity, and provider volume). Results We identified 43,997 eligible THA recipients (3% with RA) and 71,793 eligible TKA recipients (4% with RA). Total joint arthroplasty recipients with RA had higher age and sex–standardized rates of dislocation following THA (2.45%, compared with 1.21% for recipients with OA) and higher age and sex–standardized rates of infection following TKA (1.26%, compared with 0.84% for recipients with OA). Controlling for potential confounders, recipients with RA remained at increased risk of dislocation within 2 years of THA (adjusted hazard ratio [HR] 1.91, P = 0.001) and remained at increased risk of infection within 2 years of TKA (adjusted HR 1.47, P = 0.03) relative to recipients with OA. Conclusion Patients with RA are at higher risk of dislocation following THA and are at higher risk of infection following TKA relative to those with OA. Further research is warranted to elucidate explanations for these findings, including the roles of medication profile, implant choice, postoperative antibiotic protocol, and method of rehabilitation following joint replacement.

Objective To investigate the association between fast-food restaurant density and adult body mass index (BMI) in Canada. Methods Individual-level BMI and confounding variables were obtained from the 2007–2008 Canadian Community Health Survey master file. Locations of the fast-food and full-service chain restaurants and other non-chain restaurants were obtained from the 2008 Infogroup Canada business database. Food outlet density (fast-food, full-service and other) per 10,000 population was calculated for each Forward Sortation Area (FSA). Global (Moran's I) and local indicators of spatial autocorrelation of BMI were assessed. Ordinary least squares (OLS) and spatial auto-regressive error (SARE) methods were used to assess the association between local food environment and adult BMI in Canada. Results Global and local spatial autocorrelation of BMI were found in our univariate analysis. We found that OLS and SARE estimates were very similar in our multivariate models. An additional fast-food restaurant per 10,000 people at the FSA-level is associated with a 0.022 kg/m 2 increase in BMI. On the other hand, other restaurant density is negatively related to BMI. Conclusions Fast-food restaurant density is positively associated with BMI in Canada. Results suggest that restricting availability of fast-food in local neighborhoods may play a role in obesity prevention.

Objective To evaluate the safety and efficacy of corneal ultraviolet A/riboflavin collagen cross-linking (CXL) in patients with mild-to-moderate keratoconus. Design Retrospective cohort study. Participants Thirty-nine eyes from 30 patients who underwent CXL for progressive mild and moderate keratoconus were included. Methods Clinical charts for keratoconus patients who had undergone CXL alone from November 2008 to February 2011 were reviewed for preoperative and postoperative uncorrected and best corrected distance visual acuity (UDVA and BDVA, respectively), manifest refraction, topographical and Scheimpflug imaging, as well as ultrasound pachymetry and haze. Mean and steepest keratometry, central corneal thickness, and root mean-square from corneal aberrations were extracted from the topography. Central and thinnest corneal thicknesses were extracted from the Scheimpflug imaging Results A significant improvement in UDVA was seen at 3, 6 and 24 months with an average change in logMAR of 0.39 ( p = 0.003) at 24 months compared with baseline. Change in BDVA failed to reach significance, but subgroup analysis showed that a baseline BDVA worse than 0.1 logMAR was associated with better improvement. Stability of refraction, keratometry, and aberrations was demonstrated. Presence of haze was statistically significant up to 12 months ( p ≤0.001), being maximal at 3 months, then insignificant by 24 months. No complications were observed during the follow-up period. Conclusions CXL is a safe and effective stabilizing strategy for progressive mild-to-moderate keratoconus with significant improvement of the UDVA. There was a trend toward improvement of BDVA in patients with lower preoperative BDVA value.

To evaluate the use of combined topical and intracameral anesthesia for Descemet's stripping automated keratoplasty (DSAEK). This was a retrospective comparative cohort analysis consisting of 10 eyes in 10 consecutive patients undergoing DSAEK surgery with combined topical and intracameral anesthesia. These cases were compared with 21 randomly selected controls during the same time period undergoing DSAEK surgery performed under retrobulbar anesthesia. Incidence of intraoperative and postoperative complications, and endothelial cell counts were compared. In all cases, DSAEK was completed without intraoperative complications. All patients tolerated the procedure well. There were 6 cases of postoperative graft dislocation requiring rebubbling, and no cases of primary failure or endophthalmitis. No significant difference in endothelial cell counts was found at final follow-up. Short-term results suggest that combined topical and intracameral anesthesia is as safe and effective for DSAEK in cooperative patients when compared to retrobulbar anesthesia. It may thereby be an alternative anesthetic modality for patients in whom retrobulbar or peribulbar anesthesia may be risky or contraindicated.

Objective: To determine the incidence and characteristics of acute intraocular inflammation after intravitreal bevacizumab injections from a tertiary care retinal practice. Design: Retrospective cohort study. Participants: A consecutive series of patients who had received bevacizumab injections performed by a single surgeon. Methods: We reviewed the records of all patients with severe anterior chamber inflammation and (or) vitritis after bevacizumab injections. Results: A total of 693 bevacizumab injections were performed on 193 eyes of 173 patients between June 2006 and March 2008. There were a total of 9 cases of acute intraocular inflammation for an incidence of 1.30% (95% CI: 0.69%–2.47%). All patients had a worse visual acuity at the end of follow-up than on injection day. The mean loss of vision was 6.1 lines of Snellen visual acuity; one patient developed inflammation-induced glaucoma which required surgical intervention. Conclusions: Intravitreal injection of bevacizumab is associated with a low but significant risk of acute intraocular inflammation and may result in significant visual loss.

PURPOSE OF REVIEW To compare health-related quality of life (HRQoL) in American vs. Canadian populations suffering from retinal diseases. RECENT FINDINGS A search of Medline was conducted according to a strategy that combined the MeSH heading 'retinal diseases' with either of 'quality of life' as a MeSH or 'utility' as a keyword. We included studies of American or Canadian patients using any of the National Eye Institute 25-item visual functioning questionnaire (NEI-VFQ-25), visual function index (VF-14), short form health survey (SF-36), or utility to assess HRQoL. A total of 22 studies with American patients and five with Canadian patients were found. There was no significant HRQoL difference found between the American and Canadian patients for any of the HRQoL instruments (P = 0.14-0.80); however, for all instruments mean HRQoL correlated positively with mean visual acuity in the better-seeing eye (P < 0.001). SUMMARY On the basis of a review of the currently published literature, we were unable to detect significant differences in HRQoL between American and Canadian patients. Further research into the topic is necessary.

Persistent language barriers are associated with poor health outcomes. The agreement between reporting a language barrier at time of immigration and in the 2007–2008 Canadian Community Health Survey (CCHS) was calculated using kappa scores among foreign-born individuals who arrived to Ontario, Canada between 1985 and 2005. A total of 2323 immigrants were included, with a mean (± SD) time of 10.2 ± 6.4 years between immigration and completing the CCHS. Only 6 % of immigrants reported a persistent language barrier, resulting in a low agreement between the two sources (kappa = 0.06, 95 % CI 0.042–0.086). Though immigrants were less likely to report a persistent language barrier the longer they had been in Canada, only 13 % of immigrants who had arrived

Reports of bupropion misuse have increased since it was first reported in 2002. The purpose of this study was to explore trends in bupropion prescribing suggestive of misuse or diversion in Ontario, Canada.A serial cross-sectional study was conducted of Ontarians aged younger than 65 years who received prescriptions under Ontario's public drug program from April 1, 2000, to March 31, 2013. We determined the number of potentially inappropriate prescriptions in each quarter, defined as early refills dispensed within 50% of the duration of the preceding prescription, as well as potentially duplicitous prescriptions, defined as similarly early refills originating from a different prescriber and different pharmacy. We replicated these analyses for citalopram and sertraline, antidepressants not known to be prone to abuse.We identified 1,780,802 prescriptions for bupropion, 3,402,462 for citalopram, and 1,775,285 for sertraline. Rates of early refills for bupropion declined during the study from 4.8% to 3.1%. In the final quarter, rates of early refills for bupropion were more common than for citalopram (3.1% vs 2.2%) (P.001) but not for sertraline (3.1% vs 2.9%) (P =.16). Potentially duplicitous prescriptions for bupropion increased dramatically, from0.05% of all prescriptions in early 2000 to 0.47% in early 2013 and by the final quarter were more common than both citalopram (0.11%) and sertraline (0.12%) (P.001).Although no marked differences were seen for early refills of bupropion relative to its comparators, potentially duplicitous prescriptions have increased dramatically in Ontario, suggesting growing misuse of the drug.

Summary Background Suicide is the third most common cause of death among adolescents worldwide, and poisoning is the leading method of attempted suicide. Unlike more violent methods, survival after self-poisoning is common, providing an opportunity for secondary prevention. We determined the risk and time course of completed suicide after adolescent self-poisoning, and explored potential risk factors. Methods We did a population-based cohort study using multiple linked health-care databases in Ontario, Canada, from Jan 1, 2001, to Dec 31, 2012. We identified all adolescents aged 10–19 years presenting to hospital after a first self-poisoning episode. Each was matched with 50 population-based reference individuals with no such history, matching on age, sex, and year of cohort entry. The primary outcome was the risk of suicide after a first self-poisoning episode. Secondary analyses explored factors associated with suicide and self-poisoning repetition. Findings We identified 20 471 adolescents discharged from hospital after a first self-poisoning episode and 1 023 487 matched reference individuals. Over a median follow-up of 7·2 years (IQR 4·2–9·7), 248 (1%) adolescents discharged after self-poisoning died, 126 (51%) of whom died by suicide. The risk of suicide at 1 year after self-poisoning was greatly increased relative to reference individuals (hazard ratio [HR] 32·1, 95% CI 23·6–43·6), corresponding to a suicide rate of 89·6 (95% CI 75·2–106·7) per 100 000 person-years over the course of follow-up. The median time from hospital discharge to suicide was 3·0 years (IQR 1·1–5·3). Factors associated with suicide included recurrent self-poisoning (adjusted HR 3·5, 95% CI 2·4–5·0), male sex (2·5, 1·8–3·6) and psychiatric care in the preceding year (1·7, 1·1–2·5). Adolescents admitted to hospital for self-poisoning were also more likely to die from accidents (5·2, 4·1–6·6) and from all causes (3·9, 2·8–5·4) during follow-up. Interpretation Self-poisoning in adolescence is a strong predictor of suicide and premature death in the ensuing decade, and identifies a high-risk group for targeted secondary prevention. Suicide risk is increased for many years after the index hospital admission, emphasising the importance of sustained prevention efforts. Funding The Canadian Drug Safety and Effectiveness Research Network, Ontario Ministry of Health and Long-Term Care, Paediatric Consultants Partnership.

Background Cardiovascular events are responsible for half of all deaths among individuals with diabetes. Immigrants to Western countries may experience an acceleration of cardiovascular risk in the first 10 years of arrival because of a sedentary lifestyle, poor diet, or barriers to accessing care, leading to higher levels of obesity and diabetes. Objectives To compare the risk of cardiovascular events and mortality between immigrants to Canada and long-term residents with diabetes and to assess whether immigrants experience acceleration in risk after arrival. Methods We conducted a population-based retrospective cohort study using linked health and immigration data from Ontario, Canada, of 87,707 immigrants who immigrated to Canada between 1985 and 2005 matched to 87,707 long-term residents with diabetes (age ≥20 years). Individuals were followed up from April 1, 2005, until February 29, 2012, for the primary composite outcome of a cardiovascular event (acute myocardial infarction, unstable angina, congestive heart failure, transient ischemic attack, stroke) or all-cause mortality. Results There was a lower adjusted risk of cardiovascular events or mortality among immigrants (adjusted hazard ratio [HR] 0.76, 95% CI 0.74-0.78) after accounting for differences in baseline age, gender, socioeconomic status, neighborhood, and health care utilization—which persisted beyond 10 years from immigration. However, this healthy immigrant advantage was not found among more recent refugees (HR 0.93, 95% CI 0.81-1.08), immigrants with no previous education (HR 1.08, 95% CI 0.84-1.40), and those who were unmarried (HR 0.80, 95% CI 0.62-1.03). Conclusion Immigrants with diabetes are at lower risk for cardiovascular events and mortality compared with long-term residents, an effect that persists more than 10 years after arrival. Not all immigrants demonstrate this health advantage.

BACKGROUND: Dabigatran etexilate is a prodrug whose absorption is opposed by intestinal P-glycoprotein and which is converted by carboxylesterase to its active form, dabigatran. Unlike other statins, simvastatin and lovastatin are potent inhibitors of P-glycoprotein and carboxylesterase, and might either increase the risk of hemorrhage with dabigatran etexilate or decrease its effectiveness. METHODS: We conducted 2 population-based, nested case–control studies involving Ontario residents 66 years of age and older who started dabigatran etexilate between May 1, 2012, and Mar. 31, 2014. In the first study, cases were patients with ischemic stroke; in the second, cases were patients with major hemorrhage. Each case was matched with up to 4 controls by age and sex. All cases and controls received a single statin in the 60 days preceding the index date. We determined the association between each outcome and the use of simvastatin or lovastatin, relative to other statins. RESULTS: Among 45 991 patients taking dabigatran etexilate, we identified 397 cases with ischemic stroke and 1117 cases with major hemorrhage. After multivariable adjustment, use of simvastatin or lovastatin was not associated with an increased risk of stroke (adjusted odds ratio [OR] 1.33, 95% confidence interval [CI] 0.88 to 2.01). In contrast, use of simvastatin and lovastatin were associated with a higher risk of major hemorrhage (adjusted OR 1.46, 95% CI 1.17 to 1.82). INTERPRETATION: In patients receiving dabigatran etexilate, simvastatin and lovastatin were associated with a higher risk of major hemorrhage relative to other statins. Preferential use of the other statins should be considered in these patients.

Objective To determine whether the prescription of co-trimoxazole with an angiotensin converting enzyme inhibitor or angiotensin receptor blocker is associated with sudden death. Design Population based nested case-control study. Setting Ontario, Canada, from 1 April 1994 to 1 January 2012. Participants Ontario residents aged 66 years or older treated with an angiotensin converting enzyme inhibitor or angiotensin receptor blocker. Cases were those who died suddenly shortly after receiving an outpatient prescription for one of co-trimoxazole, amoxicillin, ciprofloxacin, norfloxacin, or nitrofurantoin. Each case was matched with up to four controls on age, sex, chronic kidney disease, and diabetes. Main outcome measure Odds ratio for the association between sudden death and exposure to each antibiotic relative to amoxicillin, after adjustment for predictors of sudden death according to a disease risk index. Results Of 39 879 sudden deaths, 1027 occurred within seven days of exposure to an antibiotic and were matched to 3733 controls. Relative to amoxicillin, co-trimoxazole was associated with an increased risk of sudden death (adjusted odds ratio 1.38, 95% confidence interval 1.09 to 1.76). The risk was marginally higher at 14 days (adjusted odds ratio 1.54, 1.29 to 1.84). This corresponds to approximately three sudden deaths within 14 days per 1000 co-trimoxazole prescriptions. Ciprofloxacin (a known cause of QT interval prolongation) was also associated with an increased risk of sudden death (adjusted odds ratio 1.29, 1.03 to 1.62), but no such risk was observed with nitrofurantoin or norfloxacin. Conclusions In older patients receiving angiotensin converting enzyme inhibitors or angiotensin receptor blockers, co-trimoxazole is associated with an increased risk of sudden death. Unrecognized severe hyperkalemia may underlie this finding. When appropriate, alternative antibiotics should be considered in such patients.

Background: Few studies have assessed the accuracy of administrative data for identifying multiple sclerosis (MS) patients. Objectives: To validate administrative data algorithms for MS, and describe the burden and epidemiology over time in Ontario, Canada. Methods: We employed a validated search strategy to identify all MS patients within electronic medical records, to identify patients with and without MS (reference standard). We then developed and validated different combinations of administrative data for algorithms. The most accurate algorithm was used to estimate the burden and epidemiology of MS over time. Results: The accuracy of the algorithm of one hospitalisation or five physician billings over 2 years provided both high sensitivity (84%) and positive predictive value (86%). Application of this algorithm to provincial data demonstrated an increasing cumulative burden of MS, from 13,326 patients (0.14%) in 2000 to 24,647 patients in 2010 (0.22%). Age-and-sex standardised prevalence increased from 133.9 to 207.3 MS patients per 100,000 persons in the population, from 2000 – 2010. During this same period, age-and-sex-standardised incidence varied from 17.9 to 19.4 patients per 100,000 persons. Conclusions: MS patients can be accurately identified from administrative data. Our findings illustrated a rising prevalence of MS over time. MS incidence rates also appear to be rising since 2009.

OBJECTIVE Our objective was to examine the effect of language barriers on the risk of acute and chronic complications of diabetes and on mortality among immigrants. RESEARCH DESIGN AND METHODS Linked health and immigration databases were used to identify 87,707 adults with diabetes who immigrated to Ontario, Canada, between 1985 and 2005. These individuals were included in our cohort and stratified by language ability at the time of their immigration application. Primary end points included: one or more emergency department visit or hospitalization for 1) hypo- or hyperglycemia, skin and soft tissue infection, or foot ulcer and 2) a cardiovascular event or death between April 1, 2005, and February 29, 2012. RESULTS Our cohort was followed up for a median of 6.9 person-years. Immigrants with language barriers were older (mean age, 49 ± 15 vs. 42 ± 13 years; P < 0.001), more likely to have immigrated for family reunification (66% vs. 38%, P < 0.001), had less education (secondary school or less and no education, 82% vs. 53%; P < 0.001), and a higher use of health care (mean visits, 8.6 ± 12.1 vs. 7.8 ± 11.2; P < 0.001). Immigrants with language barriers were not found to have higher adjusted rates of diabetes complications (acute complications: hazard ratio [HR] 0.99, 95% CI 0.93–1.05; cardiovascular events or death: HR 0.95, 95% CI 0.91–0.99). Significant predictors included older age, being unmarried, living in a rural neighborhood, and having less education. Immigrants who were older (≥65 years) and who had arrived through family reunification had a lower risk of cardiovascular events or death (HR 0.88, 95% CI 0.81–0.96). CONCLUSIONS In a heterogenous immigrant population with universal insurance, language barriers were not found to increase the risk of diabetes complications. However, their effect may vary based on age at time of landing, education level, marital status, and neighborhood of settlement.

David Juurlink and colleagues evaluated the risk of hospitalization with hypomagnesemia among patients taking proton pump inhibitors. Please see later in the article for the Editors' Summary, Background Some evidence suggests that proton pump inhibitors (PPIs) are an under-appreciated risk factor for hypomagnesemia. Whether hospitalization with hypomagnesemia is associated with use of PPIs is unknown. Methods and Findings We conducted a population-based case-control study of multiple health care databases in Ontario, Canada, from April 2002 to March 2012. Patients who were enrolled as cases were Ontarians aged 66 years or older hospitalized with hypomagnesemia. For each individual enrolled as a case, we identified up to four individuals as controls matched on age, sex, kidney disease, and use of various diuretic classes. Exposure to PPIs was categorized according to the most proximate prescription prior to the index date as current (within 90 days), recent (within 91 to 180 days), or remote (within 181 to 365 days). We used conditional logistic regression to estimate the odds ratio for the association of outpatient PPI use and hospitalization with hypomagnesemia. To test the specificity of our findings we examined use of histamine H2 receptor antagonists, drugs with no causal link to hypomagnesemia. We studied 366 patients hospitalized with hypomagnesemia and 1,464 matched controls. Current PPI use was associated with a 43% increased risk of hypomagnesemia (adjusted odds ratio, 1.43; 95% CI 1.06–1.93). In a stratified analysis, the risk was particularly increased among patients receiving diuretics, (adjusted odds ratio, 1.73; 95% CI 1.11–2.70) and not significant among patients not receiving diuretics (adjusted odds ratio, 1.25; 95% CI 0.81–1.91). We estimate that one excess hospitalization with hypomagnesemia will occur among 76,591 outpatients treated with a PPI for 90 days. Hospitalization with hypomagnesemia was not associated with the use of histamine H2 receptor antagonists (adjusted odds ratio 1.06; 95% CI 0.54–2.06). Limitations of this study include a lack of access to serum magnesium levels, uncertainty regarding diagnostic coding of hypomagnesemia, and generalizability of our findings to younger patients. Conclusions PPIs are associated with a small increased risk of hospitalization with hypomagnesemia among patients also receiving diuretics. Physicians should be aware of this association, particularly for patients with hypomagnesemia. Please see later in the article for the Editors' Summary, Editors' Summary Background To extract nutrients from food, we rely on a multi-stage process called digestion. A crucial stage in digestion occurs in the stomach where gastric juice, a mixture of mainly hydrochloric acid and the enzyme pepsin, breaks down the proteins present in food. We could not digest food without gastric juice, but the acid it contains, which is made by glands in the stomach, can damage the lining of the digestive system and cause symptoms of indigestion (dyspepsia), stomach (peptic) ulcers, and gastroesophageal reflux disease (GERD), a condition in which acid from the stomach leaks back up the esophagus (gullet), Acid-related disorders are often treated with proton pump inhibitors (PPIs), a class of drugs that reduces acid production in the stomach. Omeprazole, lansoprazole, and other PPIs are among the most widely prescribed drugs in the world. In 2010, 147 million prescriptions for PPIs were dispensed in the US alone. Why Was This Study Done? Like all drugs, PPIs have some unwanted side effects. They sometimes cause diarrhea, for example, and their long-term use is associated with fractures. In addition, long-term PPI use may be a risk factor for hypomagnesemia, a condition in which the magnesium level in the blood is abnormally low. If severe, hypomagnesemia can lead to life-threatening heart arrhythmias and seizures (fits). Magnesium levels are controlled by absorption of magnesium by the intestines and excretion of magnesium by the kidneys. It is thought that PPI-related hypomagnesemia involves inhibition of magnesium absorption. Given the widespread use of PPIs, it is important to know whether PPIs are a risk factor for hypomagnesemia in routine clinical practice. In this population-based case-control study, the researchers ask whether hospitalization with hypomagnesemia is associated with the use of PPIs. A case-control study compares the characteristics of individuals with a specific condition with those of matched controls without the condition. What Did the Researchers Do and Find? The researchers identified everyone aged 66 years or older who received a diagnosis of hypomagnesemia following hospital admission in Ontario over a 10 year period (366 cases) by searching a large database of hospital admissions. They identified up to four control patients from the general population who were matched with these case patients on age, sex, kidney disease, and the use of various diuretic classes (diuretic use is also associated with hypomagnesemia), and obtained data on PPI use by all patients from a database that records the prescription drugs dispensed to elderly Ontario residents. The researchers then used statistical methods to look for associations between current PPI use (a prescription within 90 days of the index date) and hospitalization with hypomagnesemia. After allowing for other characteristics that increase the risk of hypomagnesemia (including other illnesses), current PPI use was associated with a 43% increased risk of hypomagnesemia. Among patients receiving diuretics, PPI use increased the risk of hypomagnesemia by 73% whereas among patients not receiving diuretics, PPI use did not significantly increase the risk of hypomagnesemia. Finally, the researchers calculated that 76,591 individuals would need to be treated with a PPI as an outpatient for 90 days to result in one additional hospitalization with hypomagnesemia. What Do These Findings Mean? These findings show that, among elderly individuals, current (but not previous) outpatient use of PPIs is associated with an increased risk of detection of hypomagnesemia during hospitalization, particularly among patients also taking diuretics. Because this study only considered elderly patients, these findings may not apply to younger patients. Moreover, the accuracy of these findings may be affected by the validity of the hospital coding for hypomagnesemia in the database used to identify cases. Importantly, given the large number of patients that need to take PPIs to result in one additional hospitalization with hypomagnesemia, these findings should not discourage clinicians from prescribing PPIs to appropriate patients nor should they lead to calls for routine screening of magnesium levels in patients taking PPIs. Rather, these findings highlight the need for clinicians to be aware of the association between PPI use and the risk of hypomagnesemia and to reassess ongoing therapy in patients who develop hypomagnesemia while taking PPIs. Additional Information Please access these websites via the online version of this summary at http://dx.doi.org/10.1371/journal.pmed.1001736. The UK National Health Service Choices website provides information about symptoms, causes, and treatment of indigestion, heartburn and gastroesophageal reflux disease, and stomach ulcers; a “Behind the Headlines” article from 2010 discusses an editorial about the possible over-use of PPIs MedlinePlus provides links to information about indigestion, stomach ulcer, and gastroesophageal reflux disease (in English and Spanish); the MedlinePlus encyclopedia has pages on proton pump inhibitors (in English and Spanish) and on hypomagnesemia (in English and Spanish) A US Federal Drug Agency warning about the possible association between proton pump inhibitors and hypomagnesemia is available Wikipedia pages on proton pump inhibitors and on hypomagnesemia are also available (note that Wikipedia is a free online encyclopedia that anyone can edit; available in several languages)

To determine the relationship between surgeon experience with, and complications following, total joint arthroplasty (TJA) in patients with rheumatoid arthritis (RA).Using administrative data, we assembled a cohort of patients with RA who had undergone at least 1 elective primary hip or knee replacement procedure between 2002 and 2009. Cox proportional hazards, censored on death and accounting for clustering of patients within surgeons, were used to determine the relationship between overall and "RA-specific" surgeon TJA volume and the occurrence of a composite "complication" outcome (revision, infection, dislocation, or periprosthetic fracture within 2 years of the initial TJA), controlling for potential confounders (patient age, sex, comorbidity, and disease severity).We identified 4,762 patients with RA who were eligible for TJAs (1,515 total hip arthroplasties and 3,247 total knee arthroplasties). Among these patients, 152 (3.2%) experienced a surgical complication within 2 years of the procedure. After controlling for patient and hospital factors, greater surgeon TJA volume in patients with RA (RA TJA), but not overall TJA volume (all TJA), was associated with a reduced risk of complications (for surgeon RA TJA volume per 10 cases, adjusted hazard ratio [HR] 0.81, 95% confidence interval [95% CI] 0.71-0.93, P = 0.002; for surgeon all TJA volume, adjusted HR 0.98, 95% CI 0.97-1.00, P = 0.09).In a cohort of patients with RA who underwent hip or knee TJA, increased surgeon experience performing TJA in patients with RA, irrespective of their overall TJA experience and hospital factors, was associated with a decreased risk of surgical complications. These findings have potential implications for surgeon training and the referral practices of rheumatologists.

Importance Glaucoma is the second leading cause of blindness worldwide, and its insidious onset is often associated with diagnostic delay. Since glaucoma progression can often be effectively diminished when treated, identifying individuals at risk for glaucoma could potentially lead to earlier detection and prevent associated vision loss. Objective To quantify the diagnostic accuracy of examination findings and relevant risk factors in identifying individuals with primary open-angle glaucoma (POAG), the most common form of glaucoma in North America. Data Sources Structured Medline (January 1950-January 2013) search and a hand search of references and citations of retrieved articles yielding 57 articles from 41 studies. Study Selection Population-based studies of high-level methods relating relevant examination findings of cup-to-disc ratio (CDR), CDR asymmetry, intraocular pressure (IOP), and demographic risk factors to the presence of POAG. Results The summary prevalence of glaucoma in the highest-quality studies was 2.6% (95% CI, 2.1%-3.1%). Among risk factors evaluated, high myopia (≥6 diopters; odds ratio [OR], 5.7; 95% CI, 3.1-11) and family history (OR, 3.3; 95% CI, 2.0-5.6) had the strongest association with glaucoma. Black race (OR, 2.9; 95% CI, 1.4-5.9) and increasing age (especially age >80 years; OR, 2.9; 95% CI, 1.9-4.3) were also associated with an increased risk. As CDR increased, the likelihood for POAG increased with a likelihood ratio (LR) of 14 (95% CI, 5.3-39) for CDR of 0.7 or greater. Increasing CDR asymmetry was also associated with an increased likelihood for POAG (CDR asymmetry ≥0.3; LR, 7.3; 95% CI, 3.3-16). No single threshold for CDR or asymmetry ruled out glaucoma. The presence of a disc hemorrhage (LR, 12; 95% CI, 2.9-48) was highly suggestive of glaucoma, but the absence of a hemorrhage was nondiagnostic (LR, 0.94; 95% CI, 0.83-0.98). At the commonly used cutoff for high IOP (≥22), the LR was 13 (95% CI, 8.2-17), while lower IOP made glaucoma less likely (LR, 0.65; 95% CI, 0.55-0.76). We found no studies of screening examinations performed by generalist physicians in a routine setting. Conclusions and Relevance Individual findings of increased CDR, CDR asymmetry, disc hemorrhage, and elevated IOP, as well as demographic risk factors of family history, black race, and advanced age are associated with increased risk for POAG, but their absence does not effectively rule out POAG. The best available data support examination by an ophthalmologist as the most accurate way to detect glaucoma.

Most of the evidence regarding complications following total hip arthroplasty (THA) and total knee arthroplasty (TKA) are based on patients with osteoarthritis (OA); less is known about outcomes in rheumatoid arthritis (RA). Using a validated algorithm for identifying patients with RA, we undertook this study to compare the rates of complications among THA and TKA recipients between those with RA and those without RA.In patients who underwent a first primary elective THA or TKA between 2002 and 2009, those with RA were identified using a validated algorithm: a hospitalization with a diagnosis code for RA or 3 physician billing claims with a diagnosis code for RA, with at least 1 claim by a specialist (rheumatologist, orthopedic surgeon, or internist) in a 2-year period. Recipients with diagnostic codes suggesting an inflammatory arthritis, but not meeting RA criteria, were classified as having inflammatory arthritis. All remaining patients were deemed to have OA. Cox proportional hazards models, censored on death, were used to determine the relationship between the type of arthritis and the occurrence of specific complications, adjusting for potential confounders (age, sex, comorbidity, and provider volume).We identified 43,997 eligible THA recipients (3% with RA) and 71,793 eligible TKA recipients (4% with RA). Total joint arthroplasty recipients with RA had higher age and sex-standardized rates of dislocation following THA (2.45%, compared with 1.21% for recipients with OA) and higher age and sex-standardized rates of infection following TKA (1.26%, compared with 0.84% for recipients with OA). Controlling for potential confounders, recipients with RA remained at increased risk of dislocation within 2 years of THA (adjusted hazard ratio [HR] 1.91, P = 0.001) and remained at increased risk of infection within 2 years of TKA (adjusted HR 1.47, P = 0.03) relative to recipients with OA.Patients with RA are at higher risk of dislocation following THA and are at higher risk of infection following TKA relative to those with OA. Further research is warranted to elucidate explanations for these findings, including the roles of medication profile, implant choice, postoperative antibiotic protocol, and method of rehabilitation following joint replacement.

Objective To compare combined intrastromal corneal ring segment implantation with same-day ultraviolet-A/riboflavin corneal collagen cross-linking (ICRS-CXL) versus ICRS implantation alone in patients with corneal ectasia. Design Retrospective comparative study. Participants Sixty-six eyes from 54 patients with corneal ectasia were included in the study. The groups were composed of 32 eyes from 27 patients and 34 eyes from 27 patients for the ICRS-CXL and ICRS groups, respectively. Methods We reviewed the charts of all patients who underwent these procedures from November 2008 to February 2011 for preoperative and for up to 1 year postoperative uncorrected (UDVA) and best corrected distance visual acuity (BDVA), refraction, topographical analysis (mean and steepest keratometry [K]), as well as root mean-square (RMS) of higher-order aberrations (HOAs). Results Overall, a significant improvement was seen in both groups for UDVA, BDVA, sphere, cylinder, mean refractive spherical equivalent (MRSE), mean and steepest K, coma, spherical and total HOA at 12 months. Trefoil did not improve, and higher-order astigmatism worsened in the ICRS group ( p = 0.0466). There was no statistically significant difference between the 2 groups for visual acuity, sphere, cylinder, coma, trefoil, and spherical HOA. Outcomes were significantly more improved in the ICRS group for MRSE ( p = 0.0082), mean K ( p = 0.0021), steepest K ( p = 0.0152), and total HOAs ( p = 0.0208). No complications were observed. Conclusions ICRS-CXL and ICRS alone were both safe and effective in treating corneal ectasia. The ICRS alone group demonstrated better outcomes of MRSE, mean and steepest K, as well as total HOA.

Several studies have examined the rates of depression in caregivers of patients with debilitating neurologic diseases. This study describes the degree of burden and the prevalence of depression among individuals caring for legally blind patients. To the best of our knowledge, no prior studies in the ophthalmic literature have reported this relationship.Clinic-based, cross-sectional study.We included 522 individuals in Rampur, India, providing care to their close family relatives who were legally blind. Visual acuities varied from 20/200 in the best eye, to no light perception (NLP) in each eye.Several surveys were completed by the caregivers of patients whose sole impairment was visual, allowing us to quantitatively and selectively assess burden and depression among caregivers of blind patients.The Burden Index of Caregivers (BIC) was used to measure care burden and the Center for Epidemiologic Studies Depression scale was applied to determine depression.Caregivers of patients with NLP experience higher burden than caregivers of patients with lesser degrees of blindness. Daily hours spent on close supervision and the intensity of care-giving were the definitive factors linked to high BIC scores (P0.01). The prevalence of caregiver depression increased with degree of visual impairment from 16% in the 20/200 group to 48% in the NLP cohort (P0.01). Independently related variables for depression in caregivers were daily hours required for close supervision of the patient, intensity of care-giving, low household income, and the caregiver being the parent of a blind adult child (P0.01).Severity of blindness in patients directly correlated with burden in caregivers. More severe forms of blindness meant patients require more help with their activities of daily living and additional hours of close supervision per day, both of which increase care burden. The same 2 factors also increase risk of caregiver depression along with low household income and the caregiver being the parent of a blind adult child. Awareness of the extent of burden and depression among caregivers of blind individuals is required by vision health personnel to identify at risk caregivers and implement effective interventions and support strategies.

Objective To compare the rate of serious ocular and systemic adverse effects of intravitreal bevacizumab and ranibizumab in the treatment of a variety of eye diseases. Design Retrospective chart review. Participants Consecutive series of intravitreal injections of bevacizumab (n = 693) and ranibizumab (n = 891). Methods Medical records of all patients receiving injections in the series were retrieved. We considered the rate of both serious ocular adverse effects (e.g., acute intraocular inflammation, infectious endophthalmitis, retinal detachment, vitreous hemorrhage) and of arterial thromboembolic events that occurred within 1 month of injection. Results Subjects who received bevacizumab were 12 times more likely to develop severe intraocular inflammation following each injection than were those who received ranibizumab (OR = 11.71; 95% CI 1.5–93). The 1 case of acute intraocular inflammation following ranibizumab injection was mild and not associated with vision loss. No other serious ocular complications were noted. A trend was also noted toward an increased risk for arterial thromboembolic events in patients receiving bevacizumab, although the confidence interval was wide (OR = 4.26; 95% CI 0.44–41). Conclusions Significant concern still exists regarding the safety of off-label use of intravitreal bevacizumab. Patients receiving bevacizumab should be counselled regarding a possible increased risk for serious adverse events.

Objective: To compare adverse intraocular pressure (IOP) spikes in patients receiving intravitreal triamcinolone acetonide (IVTA) in 2 cohorts: ( i ) patients who underwent a topical prednisolone acetate trial (PAT) without incurring a short-term IOP rise, and ( ii ) control patients who did not undergo a PAT. Design: Retrospective cohort study. Participants: Charts of all patients who underwent any intravitreal injection during the study period were reviewed ( n = 1150). Methods: Patients in the PAT group received a 6-week course of prednisolone acetate 1% 4 times per day and had an IOP that did not rise above 25 mm Hg or above 8 mm Hg over the IOP in the contralateral eye. Patients undergoing a PAT and having a short-term IOP rise were not studied. Control patients did not receive a PAT. All patients received 12-20 mg of IVTA. Patients were followed for a minimum of 6 weeks and follow-up lasted for I year or until intraocular surgery or another IVTA injection was performed. Results: There were 97 patients in the PAT cohort and 75 control patients. Patients in the PAT cohort had a lower proportional rise between maximum IOP and baseline (43%) compared with controls (64%) ( p = 0.035). Patients in the PAT group also had a lower risk of incurring a 40% ( p = 0.05), 60% ( p = 0.0I8), and 100% ( p = 0.045) increase in maximum IOP (vs baseline) compared with controls and were less likely to require glaucoma filtration surgery ( p = 0.035). Conclusions: Patients undergoing a PAT who did not have a subsequent short-term IOP rise had a lower risk of severe IOP spikes after IVTA compared with those patients receiving IVTA but not having undergone a PAT.

To examine the effectiveness of pictograms in educating low-literacy patients in order to improve adherence to postoperative cataract regimens.Multicenter, single-blinded, randomized controlled trial.A group of 225 patients from across India, all below a 10th-grade education level, were divided into 3 groups of 75 patients.Each group was educated differently regarding medication use and frequency of dose. The control group was given verbal instruction only. Experimental group 1 (EG1) was taught using the pictograms in the clinic. Experimental group 2 (EG2) was taught in the same way as EG1 but was given the pictograms to take home. Each group was given three 10-point oral exams: on the operative day (Test 1); on postoperative day 7 (Test 2); and on day 28 (Test 3). During the patients' final visit, medication bottles were measured to ascertain use.Test 1 showed no significant difference in mean scores among groups. For Test 2, EG1 and EG2 scored similarly but significantly better than Control (control group, 5.77; EG1, 7.33; EG2, 7.62 ; p0.001). For Test 3, EG2 scored significantly better than Control and EG1 (control group, 4.37; EG1, 5.44; EG2, 7.17; p0.001). The only parameter significant for a higher test score was the participants' educational level. Higher test scores were significantly associated (p0.001) with greater medication consumption.Taking the pictograms home proved to be the most effective way to educate patients who had low literacy levels, and it increased adherence to regimens by 28 days or more. Education through pictograms strictly in the clinic was sufficient for short regimens (≤ 7 days).

Importance Suicide is the tenth leading cause of death in the United States, and its rate has risen by 16% in the past decade. Deliberate self-poisoning is the leading method of attempted suicide. Unlike more violent methods, which are almost universally fatal, survival following self-poisoning is common, providing an opportunity for secondary prevention. However, the long-term risk of suicide following a first episode of self-poisoning is unknown. Objective To determine the risk of suicide and mortality from other causes following a first self-poisoning episode. Design, Setting, and Participants Population-based cohort study using multiple linked health care databases. We identified all individuals with a first self-poisoning episode in Ontario, Canada, from April 1, 2002, through December 31, 2010, and followed up all surviving participants until December 31, 2011, or death, whichever occurred first. For each individual with a deliberate self-poisoning episode, we randomly selected 1 control from the same population with no such history, matched for age (within 3 months), sex, and calendar year. Main Outcomes and Measures The primary analysis examined the risk of suicide following discharge after self-poisoning. The secondary analyses explored factors associated with suicide and examined the risk of death caused by accidents or any other cause. Results We identified 65 784 patients (18 482 [28.1%] younger than 20 years) who were discharged after a first self-poisoning episode. During a median follow-up of 5.3 years (interquartile range, 3.1-7.6 years), 4176 died, including 976 (23.4%) by suicide. The risk of suicide following self-poisoning was markedly increased relative to controls (hazard ratio, 41.96; 95% CI, 27.75-63.44), corresponding to a suicide rate of 278 vs 7 per 100 000 person-years, respectively. The median time from hospital discharge to completed suicide was 585 days (interquartile range, 147-1301 days). Older age, male sex, multiple intervening self-poisoning episodes, higher socioeconomic status, depression, and recent psychiatric care were strongly associated with suicide. Patients with a self-poisoning episode were also more likely to die because of accidents (hazard ratio, 10.45; 95% CI, 8.10-13.47) and all causes combined (hazard ratio, 5.55; 95% CI, 5.12-6.02). Conclusions and Relevance A first self-poisoning episode is a strong predictor of subsequent suicide and premature death. Most suicides occur long after the index poisoning, emphasizing the importance of longitudinal, sustained secondary prevention initiatives.