Estimating seasonal onsets and peaks of bronchiolitis with spatially and temporally uncertain data.

RSV bronchiolitis (an acute lower respiratory tract viral infection in infants) is the most common cause of infant hospitalizations in the United States (US). The only preventive intervention currently available is monthly injections of immunoprophylaxis. However, this treatment is expensive and needs to be administered simultaneously with seasonal bronchiolitis cycles in order to be effective. To increase our understanding of bronchiolitis timing, this research focuses on identifying seasonal bronchiolitis cycles (start times, peaks, and declinations) throughout the continental US using data on infant bronchiolitis cases from the US Military Health System Data Repository. Because this data involved highly personal information, the bronchiolitis dates in the dataset were "jittered" in the sense that the recorded dates were randomized within a time window of the true date. Hence, we develop a statistical change point model that estimates spatially varying seasonal bronchiolitis cycles while accounting for the purposefully introduced jittering in the data. Additionally, by including temperature and humidity data as regressors, we identify a relationship between bronchiolitis seasonality and climate. We found that, in general, bronchiolitis seasons begin earlier and are longer in the southeastern states compared to the western states with peak times lasting approximately 1 month nationwide.

Seasonal patterns of Asthma medication fills among diverse populations of the United States.

OBJECTIVE: Nonadherence to controller and overuse of reliever asthma medications are associated with exacerbations. We aimed to determine patterns of seasonal asthma medication use and to identify time period(s) during which interventions to improve medication adherence could reduce asthma morbidity. METHODS: We conducted a retrospective cohort study of asthmatics 4-50 years of age and enrolled in three diverse health insurance plans. Seasonal patterns of medications were reported by monthly prescription fill rates per 1000 individuals with asthma from 1998 to 2013, and stratified by healthcare plan, sex, and age. RESULTS: There was a distinct and consistent seasonal fill pattern for all asthma medications. The lowest fill rate was observed in the month of July. Fills increased in the autumn and remained high throughout the winter and spring. Compared with the month of May with high medication fills, July represented a relative decrease of fills ranging from 13% (rate ratio, RR: 0.87, 95% confidence interval, 95%CI: 0.72-1.04) for the combination of inhaled corticosteroids (ICS) + long acting beta agonists (LABA) to 45% (RR: 0.55, 95%CI: 0.49-0.61) for oral corticosteroids. Such a seasonal pattern was observed each year across the 16-year study period, among healthcare plans, sexes, and ages. LABA containing control medication (ICS+LABA and LABA) fill rates were more prevalent in older asthmatics, while leukotriene receptor antagonists were more prevalent in the younger population. CONCLUSIONS: A seasonal pattern of asthma medication fill rates likely represents a reactive response to a loss of disease control and increased symptoms. Adherence to and consistent use of asthma medications among individuals who use medications in reaction to seasonal exacerbations might be a key component in reducing the risk of asthma exacerbations.

New mental health diagnoses in parents of infants admitted to a neonatal intensive care unit-a retrospective review of the Military Health System database.

OBJECTIVE: Studies suggest that parents of NICU infants are at increased risk of mental health disorders. We sought to characterize this risk using a large database. STUDY DESIGN: The Military Health System was used to retrospectively link records between parents and infants admitted to a NICU over 5 years and were matched to similar families without NICU exposure. The total study population included 35,012 infants. Logistic regression was used to estimate the association between NICU exposure and parental mental health diagnoses within 5 years of infant birth. RESULTS: Maternal NICU exposure was associated with incident diagnoses of depression (OR: 1.18-1.27, p < 0.0001), anxiety (OR: 1.06-1.18, p = 0.0151), alcohol/opiate dependence (OR: 1.29-1.52, p = 0.0079), and adjustment disorder (OR: 0.97-1.18, p = 0.0224). Paternal NICU exposure was associated with alcohol/opiate dependence (OR: 0.78-1.42, p = 0.0339). CONCLUSION: Parents of NICU infants are at risk of developing mental health disorders. Future work should identify characteristics that predict highest risk to develop effective interventions.

Comorbidities and Prescribed Medications in Patients With or Without Dry Eye Disease: A Population-Based Study.

PURPOSE: To assess the proportion of comorbidities in patients with dry eye disease (DED) compared with matched patients without DED in a comprehensive US population. DESIGN: Retrospective case-control study. METHODS: Healthcare records for insurance claims data, detailing medical services incurred by military personnel and their families and dependents in military and civilian facilities across the United States from January 1, 2003, to March 31, 2015, were obtained from the Department of Defense (DOD) Military Health System (MHS). Diagnostic and procedural codes related to DED from selected International Classification of Diseases, Ninth Revision (ICD-9) Current Procedural Terminology codes and prescriptions for cyclosporine A ophthalmic emulsion were used to identify patients with newly diagnosed and prevalent DED in the MHS database. Age, sex, and geographically matched patients without DED were also identified from healthcare claims records. Medication use and comorbidities in these patient populations were assessed and compared. RESULTS: In both the newly diagnosed and prevalent DED samples, the most common comorbidities were hypertension, cataracts, thyroid disease, type 2 diabetes, and glaucoma. All comorbidities were significantly higher in the DED vs non-DED groups (P < .001). Medication use (including, but not limited to, ophthalmic agents and drugs to treat comorbidities) was also significantly higher in the DED than in the non-DED groups (P < .001). CONCLUSIONS: The high proportions of patients with DED with a range of comorbidities and prescribed medications highlight the need for a multidisciplinary approach to the management of these patients.

Dry eye disease ranking among common reasons for seeking eye care in a large US claims database.

OBJECTIVES: Dry eye disease (DED) is a complex multifactorial condition of the ocular surface characterized by symptoms of ocular discomfort, irritation, and visual disturbance. Data previously reported from this study showed an increase in prevalence and incidence of DED with age and over time. The objective of this study was to compare the ranking of DED prevalence among other ocular conditions that led patients to seek eye care. METHODS: In this population-based study using the US Department of Defense Military Health System claims database of >9.7 million beneficiaries, indicators of DED and other ocular conditions were analyzed over time. The overall prevalence (2003-2015) and annual incidence (2008-2012) of DED and other ocular conditions were estimated using an algorithm based on two independent indicators derived from selected diagnostic and procedure codes and prescriptions for cyclosporine ophthalmic emulsion for DED and diagnostic codes for the indicators of other common ocular conditions. RESULTS: In 2003-2015, the most common ocular conditions were disorders of refraction and accommodation (25.84%), cataracts (17.14%), glaucoma (7.27%), disorders of the conjunctiva (6.76%), other retinal disorders (5.94%), and DED (5.28%). DED was the fifth most prevalent ocular condition in women (7.78%) and ninth most prevalent in men (2.96%). In 2012, DED had the third highest annual incidence (0.87%), behind disorders of refraction/accommodation (1.87%) and cataracts (1.50%). CONCLUSION: This study provided further epidemiologic evidence for DED as a commonly occurring condition that drives patients to seek treatment.

Estimated Prevalence and Incidence of Dry Eye Disease Based on Coding Analysis of a Large, All-age United States Health Care System.

PURPOSE: To assess overall prevalence, annual prevalence, and incidence of dry eye disease (DED) in a large, representative population in the United States. DESIGN: Prevalence and incidence study. METHODS: Retrospective analysis using the Department of Defense (DOD) Military Health System (MHS) data on beneficiary medical claims from United States DOD military and civilian facilities, January 1, 2003 through March 31, 2015. PATIENT POPULATION: Using an algorithm, medical diagnostic codes indicative of DED and prescriptions for cyclosporine ophthalmic emulsion identified a DED population from 9.7 million MHS beneficiaries (DOD service members, retirees, and dependents, aged 2-80+ years). MAIN OUTCOME MEASURES: DED overall prevalence (2003-2015), annual prevalence (2005-2012), and annual incidence (2008-2012) stratified by sex, age group, and International Statistical Classification of Diseases and Related Health Problems, Ninth Revision diagnosis code grouping. RESULTS: DED prevalence was 5.28% overall, 7.78% among female beneficiaries, 2.96% among male beneficiaries and increased with age from 0.20% for ages 2-17 years, to 11.66% for individuals aged 50+ years. Annual prevalence increased from 0.8% to 3.0% overall, from 1.4% to 4.5% in female beneficiaries, and from 0.3% to 1.6% in male beneficiaries. Annual prevalence increased across age groups starting at age 18-39, 0.1%-0.6%, to age 50+, 1.8%-6.0%. Annual incidence increased from 0.6% to 0.9% overall, from 0.8% to 1.2% in female beneficiaries, and from 0.3% to 0.6% in male beneficiaries. Across age groups, annual incidence increased starting at age 18-39 (0.2%-0.3%), to age 50+ (1.0%-1.6%). CONCLUSIONS: DED overall prevalence, annual prevalence, and incidence were found to increase over time for all demographics. These findings highlight the continued importance of research and therapeutic development for this common condition.

Repeat Hospitalizations Predict Mortality in Patients With Heart Failure.

INTRODUCTION: Heart failure (HF) affects more than 5.1 million Americans and is projected to increase. Understanding the relationship between hospitalization and mortality can help to guide clinical management. The aim of the study is to evaluate the impact of repeat HF hospitalizations on all-cause mortality and to determine risk variables related to patient mortality. MATERIALS AND METHODS: Using administrative data from the Military Health System, a cohort of patients with an index admission for HF between 2007 and 2011 was identified. HF hospitalizations were defined as any hospital claim with an International Classification of Diseases, Ninth Revision diagnosis of 428.xx in the primary diagnosis field over the 7-year study period (2007-2013). Patients were subsequently categorized based on total number of HF hospitalizations. A multivariate Cox regression model, adjusting for age, sex, and comorbidities, was used to estimate hazard ratios. Kaplan-Meier survival curves were constructed based on the frequency of HF hospitalizations. RESULTS: Of the 51,286 patients admitted for HF, 54.7% were male with a mean (SD) age of 76.3 (10.8) years, and 29,714 died during 135,211 person-years of follow-up. Mean survival time was 2.6, 1.8, 1.5, and 1.3 years after the first, second, third, and fourth hospitalization, respectively. The mortality rate of patients at 30 days and 1 year postindex HF hospitalization was 7.4% and 27.3%, respectively. A history of dementia and chronic kidney disease without dialysis decreased overall survival. CONCLUSIONS: Repeat HF hospitalizations remain a strong predictor of mortality for existing patients with HF. As a result, clinicians and patients can individualize the optimal treatment strategy and resources on the basis of the suspected prognosis.

Predictors of Revision Surgery After Anterior Cruciate Ligament Reconstruction.

BACKGROUND: Arthroscopically assisted anterior cruciate ligament (ACL) reconstruction is a common orthopaedic procedure. Graft failure after reconstruction remains a devastating complication, often requiring revision surgery and less aggressive or modified rehabilitation. Worse functional and patient-reported outcomes are reported compared with primary reconstruction. Moreover, both rates and risk factors for revision are variable and inconsistent within the literature. PURPOSE: To determine the rate of revision surgery after ACL reconstruction in a large cohort of patients, to assess the influence of patient characteristics on the odds of revision, and to compare revision rates between active-duty military members and non-active-duty beneficiaries. STUDY DESIGN: Descriptive epidemiology study. METHODS: Using administrative data from the Military Health System, a retrospective study was designed to characterize the rate of ACL revision surgery among patients treated within a military facility. All patients ≥18 years at the time of ACL reconstruction were identified using the American Medical Association Current Procedural Terminology (CPT) for ACL reconstruction (CPT code 29888) over 7 years (2005-2011). Revision ACL reconstructions were identified as having ≥2 ACL reconstruction procedure codes on the ipsilateral knee at least 90 days apart. Univariate analysis was performed to calculate odds ratios (ORs) for demographic, perioperative medication use, and concomitant procedure-related risk factors. A multivariate logistic regression model determined risk covariates in the active-duty cohort. RESULTS: The study population consisted of 17,164 ACL reconstructions performed among 16,336 patients, of whom 83.3% were male with a mean ± SD age of 28.9 ± 7.6 years for the nonrevision group, and was predominantly active duty (89.2%). Patients undergoing ACL reconstruction on both knees only contributed their index knee for analyses. There were 587 patients who underwent revision surgery, corresponding to an overall revision rate of 3.6%. The median time from the index surgery to revision surgery was 500 days (interquartile range, 102-2406 days). Revision rates were higher in the active-duty cohort as compared with non-active-duty beneficiaries (3.8% vs 1.8%, respectively; OR, 2.14; 95% CI, 1.49-3.07). Based on multivariate logistic regression in the active-duty cohort, age ≥35 years (OR, 0.44; 95% CI, 0.33-0.58) and concomitant meniscal repair (OR, 0.69; 95% CI, 0.53-0.91) were found to be protective with regard to the odds of revision surgery. Perioperative medication use of nonsteroidal anti-inflammatory drugs (NSAIDs) (OR, 1.33; 95% CI, 1.12-1.58; number needed to harm [NNH], 100) and COX-2 inhibitors (OR, 1.31; 95% CI, 1.04-1.66; NNH, 333) was associated with increased odds of revision surgery. No significant findings were detected among sex, race, nicotine use, body mass index, or other concomitant procedures of interest. CONCLUSION: In this large cohort study, the rate of revision ACL reconstruction was 3.6%, which is consistent with the existing literature. Increased odds of revision surgery among active-duty personnel were associated with the perioperative use of NSAIDs and COX-2 inhibitors. Age ≥35 years and concomitant meniscal repair were found to be protective against ACL revision.

The Prevalence of Symptomatic Deep Venous Thrombosis and Pulmonary Embolism After Anterior Cruciate Ligament Reconstruction.

BACKGROUND: Arthroscopically assisted anterior cruciate ligament (ACL) reconstruction is a common orthopaedic procedure. The incidence and risk factors of venous thromboembolism (VTE) after ACL reconstruction remain unclear. PURPOSE: To define the incidence of VTE after ACL reconstruction and identify associated risk factors in a large cohort of patients. STUDY DESIGN: Descriptive epidemiological study. METHODS: All patients aged ≥18 years who underwent ACL reconstruction between 2005 and 2011 were identified from the Department of Defense Medical Data Repository. The prevalence of VTE, including deep venous thrombosis (DVT) and pulmonary embolism (PE), within 3 months of ACL reconstruction was queried. Univariate analyses were performed to define odds ratios (ORs) for demographic, medication use, and procedural-related risk factors. RESULTS: A total of 87 VTE events (0.53% [95% CI, 0.42%-0.65%]) occurred after 16,558 ACL reconstructions performed on 15,767 patients. DVT was documented after 55 procedures and PE after 35 procedures. Three patients were documented to have both DVT and PE within the study period. The odds of VTE increased in patients aged ≥35 years (OR, 1.96 [95% CI, 1.27-3.04]; P = .003). Nicotine history increased the odds of DVT (OR, 1.99 [95% CI, 1.15-3.43]; P = .014). Concomitant high tibial osteotomy (HTO) increased the odds of PE (OR, 18.31 [95% CI, 2.4-139.6]; P = .005), whereas concomitant posterior cruciate ligament (PCL) reconstruction increased the odds of both VTE (OR, 3.43 [95% CI, 1.07-11.2]; P = .38) and DVT (OR, 5.57 [95% CI, 1.71-18.14]; P = .004). Nonsteroidal drug use was associated with decreased odds for VTE and DVT (OR, 0.44 [95% CI, 0.28-0.70]; P < .001 and OR, 0.38 [95% CI, 0.22-0.69]; P < .001, respectively). Anticoagulants were associated with increased odds for VTE, DVT, and PE (OR, 98.32 [95% CI, 61.63-156.86]; P < .001; OR, 111.93 [95% CI, 63.95-195.92]; P < .001; and OR, 47.84 [95% CI, 22.55-101.52]; P < .001, respectively). No detectible difference in odds was found for sex, body mass index, or aspirin or cyclooxygenase-2 inhibitor use. CONCLUSION: The incidence of VTE after ACL reconstruction in this large population was low. Increased odds of VTE was identified in patients aged ≥35 years with a history of nicotine use, anticoagulant use, concomitant HTO, or concomitant PCL reconstruction. Controlled studies are necessary to determine the efficacy of chemoprophylaxis and to develop evidence-based clinical practice guidelines to minimize VTE after ACL reconstruction.