Associations between Missed Colonoscopy Appointments and Multiple Prior Adherence Behaviors in an Integrated Healthcare System: An Observational Study.

BACKGROUND: Missed colonoscopy appointments delay screening and treatment for gastrointestinal disorders. Prior nonadherence with other care components may be associated with missed colonoscopy appointments. OBJECTIVE: To assess variability in prior adherence behaviors and their association with missed colonoscopy appointments. DESIGN: Retrospective cohort study. PARTICIPANTS: Patients scheduled for colonoscopy in an integrated healthcare system between January 2016 and December 2018. MAIN MEASURES: Prior adherence behaviors included: any missed outpatient appointment in the previous year; any missed gastroenterology clinic or colonoscopy appointment in the previous 2 years; and not obtaining a bowel preparation kit pre-colonoscopy. Other sociodemographic, clinical, and system characteristics were included in a multivariable model to identify independent associations between prior adherence behaviors and missed colonoscopy appointments. KEY RESULTS: The median age of the 57,590 participants was 61 years; 52.8% were female and 73.4% were white. Of 77,684 colonoscopy appointments, 3,237 (4.2%) were missed. Individuals who missed colonoscopy appointments were more likely to have missed a previous primary care appointment (62.5% vs. 38.4%), a prior gastroenterology appointment (18.4% vs. 4.7%) or not to have picked up a bowel preparation kit (42.4% vs. 17.2%), all p < 0.001. Correlations between the three adherence measures were weak (phi < 0.26). The rate of missed colonoscopy appointments increased from 1.8/100 among individuals who were adherent with all three prior care components to 24.6/100 among those who were nonadherent with all three care components. All adherence variables remained independently associated with nonadherence with colonoscopy in a multivariable model that included other covariates; adjusted odds ratios (with 95% confidence intervals) were 1.6 (1.5-1.8) for outpatient appointments, 1.9 (1.7-2.1) for gastroenterology appointments, and 3.1 (2.9-3.4) for adherence with bowel preparation kits, respectively. CONCLUSIONS: Three prior adherence behaviors were independently associated with missed colonoscopy appointments. Studies to predict adherence should use multiple, complementary measures of prior adherence when available.

Update of a Multivariable Opioid Overdose Risk Prediction Model to Enhance Clinical Care for Long-term Opioid Therapy Patients.

BACKGROUND: Clinical opioid overdose risk prediction models can be useful tools to reduce the risk of overdose in patients prescribed long-term opioid therapy (LTOT). However, evolving overdose risk environments and clinical practices in addition to potential harmful model misapplications require careful assessment prior to widespread implementation into clinical care. Models may need to be tailored to meet local clinical operational needs and intended applications in practice. OBJECTIVE: To update and validate an existing opioid overdose risk model, the Kaiser Permanente Colorado Opioid Overdose (KPCOOR) Model, in patients prescribed LTOT for implementation in clinical care. DESIGN, SETTING, AND PARTICIPANTS: The retrospective cohort study consisted of 33, 625 patients prescribed LTOT between January 2015 and June 2019 at Kaiser Permanente Colorado, with follow-up through June 2021. MAIN MEASURES: The outcome consisted of fatal opioid overdoses identified from vital records and non-fatal opioid overdoses from emergency department and inpatient settings. Predictors included demographics, medication dispensings, substance use disorder history, mental health history, and medical diagnoses. Cox proportional hazards regressions were used to model 2-year overdose risk. KEY RESULTS: During follow-up, 65 incident opioid overdoses were observed (111.4 overdoses per 100,000 person-years) in the study cohort, of which 11 were fatal. The optimal risk model needed to risk-stratify patients and to be easily interpreted by clinicians. The original 5-variable model re-validated on the new study cohort had a bootstrap-corrected C-statistic of 0.73 (95% CI, 0.64-0.85) compared to a C-statistic of 0.80 (95% CI, 0.70-0.88) in the updated model and 0.77 (95% CI, 0.66-0.87) in the final adapted 7-variable model, which was also well-calibrated. CONCLUSIONS: Updating and adapting predictors for opioid overdose in the KPCOOR Model with input from clinical partners resulted in a parsimonious and clinically relevant model that was poised for integration in clinical care.

Telemedicine-based inspiratory muscle training and walking promotion with lung cancer survivors following curative intent therapy: a parallel-group pilot randomized trial.

PURPOSE: Following curative-intent therapy of lung cancer, many survivors experience dyspnea and physical inactivity. We investigated the feasibility, acceptability, safety, and potential efficacy of inspiratory muscle training (IMT) and walking promotion to disrupt a postulated "dyspnea-inactivity" spiral. METHODS: Between January and December 2022, we recruited lung cancer survivors from Kaiser Permanente Colorado who completed curative-intent therapy within 1-6 months into a phase-IIb, parallel-group, pilot randomized trial (1:1 allocation). The 12-week intervention, delivered via telemedicine, consisted of exercise training (IMT + walking), education, and behavior change support. Control participants received educational materials on general exercise. We determined feasibility a priori: enrollment of ≥ 20% eligible patients, ≥ 75% retention, study measure completion, and adherence. We assessed acceptability using the Telemedicine-Satisfaction-and-Usefulness-Questionnaire and safety events that included emergency department visits or hospitalizations. Patient-centered outcome measures (PCOMs) included dyspnea (University-of-California-San-Diego-Shortness-of-Breath-Questionnaire), physical activity (activPAL™ steps/day), functional exercise capacity (mobile-based-six-minute-walk-test), and health-related quality of life (HRQL, St.-George's-Respiratory-Questionnaire). We used linear mixed-effects models to assess potential efficacy. RESULTS: We screened 751 patients, identified 124 eligible, and consented 31 (25%) participants. Among 28 participants randomized (14/group), 22 (11/group) completed the study (79% retention). Intervention participants returned > 90% of self-reported activity logs, completed > 90% of PCOMs, and attended > 90% of tele-visits; 75% of participants performed IMT at the recommended dose. Participants had high satisfaction with tele-visits and found the intervention useful. There was no statistically significant difference in safety events between groups. Compared to control participants from baseline to follow-up, intervention participants had statistically significant and clinically meaningful improved HRQL (SGRQ total, symptom, and impact scores) (standardized effect size: -1.03 to -1.30). CONCLUSIONS: Among lung cancer survivors following curative-intent therapy, telemedicine-based IMT + walking was feasible, acceptable, safe, and had potential to disrupt the "dyspnea-inactivity" spiral. Future efficacy/effectiveness trials are warranted and should incorporate IMT and walking promotion to improve HRQL. TRIAL REGISTRATION: ClinicalTrials.gov NCT05059132.

Pre-Operative Predictors for Discharge to Post-Acute Care Facilities After Total Knee Arthroplasty.

BACKGROUND: Joint replacement surgery is in increasing demand and is the most common inpatient surgery for Medicare beneficiaries. The venue for post-operative rehabilitation, including early outpatient therapy after surgery, influences recovery and quality of life. As part of a comprehensive total joint program at Kaiser Permanente Colorado, we developed and validated a predictive model to anticipate and plan the disposition for rehabilitation of our patients after total knee arthroplasty (TKA). METHODS: We analyzed data for TKA patients who completed a pre-operative Total Knee Risk Assessment in 2017 (the model development cohort) or during the first 6 months of 2018 (the model validation cohort). The Total Knee Risk Assessment, which is used to guide disposition for rehabilitation, included questions in mobility, social, and environment domains. Multivariable logistic regression was used to predict discharge to post-acute care facilities (PACFs) (ie, skilled nursing facilities or acute rehabilitation centers). RESULTS: Data for a total of 1481 and 631 patients who underwent TKA were analyzed in the development and validation cohorts, respectively. Ninety-three patients (6.3%) in the development cohort and 22 patients (3.5%) in the validation cohort were discharged to PACFs. Eight risk factors for discharge to PACFs were included in the final multivariable model. Patients with a diagnosis of neurological disorder and with a mobility/balance issue had the greatest chance of discharge to PACFs. CONCLUSION: This validated predictive model for discharge disposition following TKA may be used as a tool in shared decision-making and discharge planning for patients undergoing TKA.

Enhancing effect of macroporous adsorption resin on gamma-aminobutyric acid production by Enterococcus faecium in whole-cell biotransformation system.

Gamma-aminobutyric acid (GABA) biosynthesis depended to a great extent on the biotransformation characterization of glutamate decarboxylase (GAD) and process conditions. In this paper, the enhancing effect of D101 macroporous adsorption resin (MAR) on the GABA production was investigated based on the whole-cell biotransformation characterization of Enterococcus faecium and adsorption characteristics of D101 MAR. The results indicated that the optimal pH for reaction activity of whole-cell GAD and pure GAD was 4.4 and 5.0, respectively, and the pH range retained at least 50% of GAD activity was from 4.8 to 5.6 and 4.0-4.8, respectively. No substrate inhibition effect was observed on both pure GAD and whole-cell GAD, and the maximum activity could be obtained when the initial L-glutamic acid (L-Glu) concentration exceeded 57.6 mmol/L and 96.0 mmol/L, respectively. Besides, GABA could significantly inhibit the activity of whole-cell GAD rather than pure GAD. When the initial GABA concentration of the reaction solution remained 100 mmol/L, 33.51 ± 9.11% of the whole-cell GAD activity was inhibited. D101 MAR exhibited excellent properties in stabilizing the pH of the conversion reaction system, supplementing free L-Glu and removing excess GABA. Comparison of the biotransformation only in acetate buffer, the GABA production, with 50 g/100 mL of D101 MAR, was significantly increased by 138.71 ± 5.73%. D101 MAR with pre-adsorbed L-Glu could significantly enhance the production of GABA by gradual replenishment of free L-Glu, removing GABA and maintaining the pH of the reaction system, which would eventually make the GABA production more economical and eco-friendly.

Development and Validation of a Model for Predicting Rehabilitation Care Location Among Patients Discharged Home After Total Knee Arthroplasty.

BACKGROUND: Demand for joint replacement is increasing, with many patients receiving postsurgical physical therapy (PT) in non-inpatient settings. Clinicians need a reliable tool to guide decisions about the appropriate PT setting for patients discharged home after surgery. We developed and validated a model to predict PT location for patients in our health system discharged home after total knee arthroplasty. METHODS: We analyzed data for patients who completed a preoperative total knee risk assessment in 2017 (model development cohort) or during the first 6 months of 2018 (model validation cohort). The initial total knee risk assessment, to guide rehabilitation disposition, included 28 variables in mobility, social, and environment domains, and on patient demographics and comorbidities. Multivariable logistic regression was used to identify factors that best predict discharge to home health service (HHS) vs home with outpatient PT. Model performance was assessed by standard criteria. RESULTS: The development cohort included 259 patients (19%) discharged to HHS and 1129 patients (81%) discharged to home with outpatient PT. The validation cohort included 609 patients, with 91 (15%) discharged to HHS. The final model included age, gender, motivation for outpatient PT, and reliable transportation. Patients without motivation for outpatient PT had the highest probability of discharge to HHS, followed by those without reliable transportation. Model performance was excellent in the development and validation cohort, with c-statistics of 0.91 and 0.86, respectively. CONCLUSION: We developed and validated a predictive model for total knee arthroplasty PT discharge location. This model includes 4 variables with accurate prediction to guide patient-clinician preoperative decision making.

Risk of Breast Cancer With Long-Term Use of Calcium Channel Blockers or Angiotensin-Converting Enzyme Inhibitors Among Older Women.

Controversy exists about breast cancer risk associated with long-term use of calcium channel blockers (CCBs) or angiotensin-converting enzyme inhibitors (ACEis), respectively. Our objective in this study was to separately evaluate associations between duration of CCB or ACEi use and breast cancer in hypertensive women aged ≥55 years at 3 sites in the Kaiser Permanente health-care system (1997­2012). Exposures included CCB or ACEi use of 1­12 years' duration, determined from pharmacy dispensings. Outcomes included invasive lobular or ductal carcinoma. Statistical methods included discrete-time survival analyses. The cohort included 19,674 (17.9%) CCB users and 90,078 (82.1%) ACEi users. Two percent (n = 397) of CCB users and 1.9% (n = 1,733) of ACEi users developed breast cancer. Compared with 1­<2 years of use, in adjusted analysis, there was no association between CCB use for 2­<12 years and breast cancer: All 95% confidence intervals included 1. Increasing duration of ACEi use was associated with reduced breast cancer risk: Compared with 1­<2 years of use, the adjusted hazard ratio was 0.76 (95% confidence interval: 0.63, 0.92) for 5­<6 years of use and 0.63 (95% confidence interval: 0.43, 0.93) for 9­<10 years of use. We conclude that among older women with hypertension, long-term CCB use does not increase breast cancer risk and long-term treatment with ACEis may confer protection against breast cancer.

Effect of continuity of care on hospital utilization for seniors with multiple medical conditions in an integrated health care system.

PURPOSE: Lower continuity of care has been associated with higher rates of adverse outcomes for persons with multiple chronic medical conditions. It is unclear, however, whether this relationship also exists within integrated systems that offer high levels of informational continuity through shared electronic health records. METHODS: We conducted a retrospective cohort study of 12,200 seniors with 3 or more chronic conditions within an integrated delivery system. Continuity of care was calculated using the Continuity of Care Index, which reflects visit concentration with individual clinicians. Using Cox proportional hazards regression permitting continuity to vary monthly until the outcome or censoring event, we separately assessed inpatient admissions and emergency department visits as a function of primary care continuity and specialty care continuity. RESULTS: After adjusting for covariates (demographics; baseline, primary, and specialty care visits; baseline outcomes; and morbidity burden), greater primary care continuity and greater specialty care continuity were each associated with a lower risk of inpatient admission (respective hazard ratios (95% CIs) = 0.97 (0.96, 0.99) and 0.95 (0.93, 0.98)) and a lower risk of emergency department visits (respective hazard ratios = 0.97 (0.96, 0.98) and 0.98 (0.96, 1.00)). For the subgroup with 3 or more primary care and 3 or more specialty care visits, specialty care continuity (but not primary care continuity) was independently associated with a decreased risk of inpatient admissions (hazard ratio = 0.94 (0.92, 0.97)), and primary care continuity (but not specialty care continuity) was associated with a decreased risk of emergency department visits (hazard ratio = 0.98 (0.96, 1.00)). CONCLUSIONS: In an integrated delivery system with high informational continuity, greater continuity of care is independently associated with lower hospital utilization for seniors with multiple chronic medical conditions. Different subgroups of patients will benefit from continuity with primary and specialty care clinicians depending on their care needs.

Assessment of morbidity over time in predicting health outcomes.

BACKGROUND: Administratively derived morbidity measures are often used in observational studies as predictors of outcomes. These typically reflect a limited time period before an index event; some outcomes may be affected by rate of morbidity change over longer preindex periods. OBJECTIVES: The aim of the study was to develop statistical models representing the trajectory of individual morbidity over time and to evaluate the performance of trajectory versus other summary morbidity measures in predicting a range of health outcomes. METHODS: From a retrospective cohort study of integrated health system members aged 65 years or older with 3 or more common chronic medical conditions, we used available diagnoses for up to 10 years to examine associations between variations of the Charlson Comorbidity Index (CCI, Quan adaptation) and health outcomes. A linear mixed effects model was used to estimate the trajectory of individual CCI over time; estimated parameters describing individual trajectories were used as predictors for health outcomes. Other variations of CCI were: a "snapshot" measure, a cumulative measure, and actual baseline and rate of change. Models were developed in an initial cohort for whom we had survey data, and verified in a larger cohort. RESULTS: Among 961 surveyed members and 13,163 members of a secondary cohort, cumulative and snapshot measures provided best fit and predictive ability for utilization outcomes. Incorporating trajectory resulted in a slightly better model for self-reported health status. CONCLUSIONS: Modeling longitudinal morbidity trajectories did not add substantially to the association between morbidity and utilization or mortality. Standard snapshot morbidity measures likely sufficiently capture multimorbidity in assessing these outcomes.

Signal detection of adverse events with imperfect confirmation rates in vaccine safety studies using self-controlled case series design.

The Vaccine Safety Datalink project captures electronic health record data including vaccinations and medically attended adverse events on 8.8 million enrollees annually from participating managed care organizations in the United States. While the automated vaccination data are generally of high quality, a presumptive adverse event based on diagnosis codes in automated health care data may not be true (misclassification). Consequently, analyses using automated health care data can generate false positive results, where an association between the vaccine and outcome is incorrectly identified, as well as false negative findings, where a true association or signal is missed. We developed novel conditional Poisson regression models and fixed effects models that accommodate misclassification of adverse event outcome for self-controlled case series design. We conducted simulation studies to evaluate their performance in signal detection in vaccine safety hypotheses generating (screening) studies. We also reanalyzed four previously identified signals in a recent vaccine safety study using the newly proposed models. Our simulation studies demonstrated that (i) outcome misclassification resulted in both false positive and false negative signals in screening studies; (ii) the newly proposed models reduced both the rates of false positive and false negative signals. In reanalyses of four previously identified signals using the novel statistical models, the incidence rate ratio estimates and statistical significances were similar to those using conventional models and including only medical record review confirmed cases.

Bias correction of risk estimates in vaccine safety studies with rare adverse events using a self-controlled case series design.

The self-controlled case series (SCCS) method is often used to examine the temporal association between vaccination and adverse events using only data from patients who experienced such events. Conditional Poisson regression models are used to estimate incidence rate ratios, and these models perform well with large or medium-sized case samples. However, in some vaccine safety studies, the adverse events studied are rare and the maximum likelihood estimates may be biased. Several bias correction methods have been examined in case-control studies using conditional logistic regression, but none of these methods have been evaluated in studies using the SCCS design. In this study, we used simulations to evaluate 2 bias correction approaches-the Firth penalized maximum likelihood method and Cordeiro and McCullagh's bias reduction after maximum likelihood estimation-with small sample sizes in studies using the SCCS design. The simulations showed that the bias under the SCCS design with a small number of cases can be large and is also sensitive to a short risk period. The Firth correction method provides finite and less biased estimates than the maximum likelihood method and Cordeiro and McCullagh's method. However, limitations still exist when the risk period in the SCCS design is short relative to the entire observation period.

Ethical Aspects of Physician Decision-Making for Deprescribing Among Older Adults With Dementia.

Importance: Physicians endorse deprescribing of risky or unnecessary medications for older adults (aged ≥65 years) with dementia, but there is a lack of information on what influences decisions to deprescribe in this population. Objective: To understand how physicians make decisions to deprescribe for older adults with moderate dementia and ethical and pragmatic concerns influencing those decisions. Design, Setting, and Participants: A cross-sectional national mailed survey study of a random sample of 3000 primary care physicians from the American Medical Association Physician Masterfile who care for older adults was conducted from January 15 to December 31, 2021. Main Outcomes and Measures: The study randomized participants to consider 2 clinical scenarios in which a physician may decide to deprescribe a medication for older adults with moderate dementia: 1 in which the medication could cause an adverse drug event if continued and the other in which there is no evidence of benefit. Participants ranked 9 factors related to possible ethical and pragmatic concerns through best-worst scaling methods (from greatest barrier to smallest barrier to deprescribing). Conditional logit regression quantified the relative importance for each factor as a barrier to deprescribing. Results: A total of 890 physicians (35.0%) returned surveys; 511 (57.4%) were male, and the mean (SD) years since graduation was 26.0 (11.7). Most physicians had a primary specialty in family practice (50.4% [449 of 890]) and internal medicine (43.5% [387 of 890]). A total of 689 surveys were sufficiently complete to analyze. In both clinical scenarios, the 2 greatest barriers to deprescribing were (1) the patient or family reporting symptomatic benefit from the medication (beneficence and autonomy) and (2) the medication having been prescribed by another physician (autonomy and nonmaleficence). The least influential factor was ease of paying for the medication (justice). Conclusions and Relevance: Findings from this national survey study of primary care physicians suggests that understanding ethical aspects of physician decision-making can inform clinician education about medication management and deprescribing decisions for older adults with moderate dementia.

Association of patient-centered outcomes with patient-reported and ICD-9-based morbidity measures.

PURPOSE: Evaluating patient-centered care for complex patients requires morbidity measurement appropriate for use with a variety of clinical outcomes. We compared the contributions of self-reported morbidity and morbidity measured using administrative diagnosis data for both patient-reported outcomes and utilization outcomes. METHODS: Using a cohort of 961 persons aged 65 years or older with 3 or more medical conditions, we explored 9 health outcomes as a function of 4 independent variables representing different types of morbidity measures: International Classification of Diseases, Ninth Revision (ICD-9), a self-reported weighted count of conditions, and self-reported symptoms of depression and of anxiety. Outcomes varied from self-reported health status to utilization. Depending on the outcome measure, we used multivariate linear, negative binomial, or logistic regression, adjusting for demographic characteristics and length of enrollment to assess associations between dependent and all 4 independent variables. RESULTS: Higher morbidity measured by ICD-9 diagnoses was independently associated with less favorable levels of 7 of the 9 clinical outcomes. Higher self-reported disease burden was significantly associated with less favorable levels of 8 of the outcomes, controlling for the 3 other morbidity measures. Morbidity measured by diagnosis code was more strongly associated with higher utilization, whereas self-reported disease burden and emotional symptoms were more strongly associated with patient-reported outcomes. CONCLUSIONS: A comprehensive assessment of morbidity requires both subjective and objective measurement of disease burden as well as an assessment of emotional symptoms. Such multidimensional morbidity measurement is particularly relevant for research or quality assessments involving the delivery of patient-centered care to complex patient populations.

Patterns of beta-blocker intensification in ambulatory heart failure patients and short-term association with hospitalization.

BACKGROUND: In response to the short-term negative inotropic and chronotropic effects of ß-blockers, heart failure (HF) guidelines recommend initiating ß-blockers at low dose with gradual uptitration as tolerated to doses used in clinical trials. However, patterns and safety of ß-blocker intensification in routine practice are poorly described. METHODS: We described ß-blocker intensification among Kaiser Colorado enrollees with a primary discharge diagnosis of HF between 2001-2009. We then assessed ß-blocker intensification in the 30 days prior to first hospital readmission for cases compared to the same time period following index hospitalization for non-rehospitalized matched controls. In separate analysis of the subgroup initiated on ß-blocker after index hospital discharge, we compared adjusted rates of 30-day hospitalization following initiation of high versus low dose ß-blocker. RESULTS: Among 3,227 patients, median age was 76 years and 37% had ejection fraction ≤ 40% (LVSD). During a median follow up of 669 days, 14% were never on ß-blocker, 21% were initiated on ß-blocker, 43% were discharged on ß-blocker but never uptitrated, and 22% had discharge ß-blocker uptitrated; 63% were readmitted and 49% died. ß-blocker intensification occurred in the 30 days preceding readmission for 39 of 1,674 (2.3%) readmitted cases compared to 27 (1.6%) of matched controls (adjusted OR 1.36, 95% CI 0.81-2.27). Among patients initiated on therapy, readmission over the subsequent 30 days occurred in 6 of 155 (3.9%) prescribed high dose and 9 of 513 (1.8%) prescribed low dose ß-blocker (adjusted OR 3.10, 95% CI 1.02-9.40). For the subgroup with LVSD, findings were not significantly different. CONCLUSION: While ß-blockers were intensified in nearly half of patients following hospital discharge and high starting dose was associated with increased readmission risk, the prevailing finding was that readmission events were rarely preceded by ß-blocker intensification. These data suggest that ß-blocker intensification is not a major precipitant of hospitalization, provided recommended dosing is followed.

Association between changes in loneliness identified through screening and changes in depression or anxiety in older adults.

BACKGROUND: Changes in loneliness are associated with corresponding changes in depression, anxiety, and general health in population surveys, but few studies have assessed these associations through repeated screening in clinical settings. METHODS: Retrospective cohort study among individuals ≥age 65 in an integrated health care system who completed loneliness screening before two annual wellness visits, separated by a mean of 12.9 (SD 2.0) months, between 2013 and 2018. Their responses identified four subgroups: individuals who were persistently lonely; not lonely; experienced an increase (recently lonely); or decrease (previously lonely) in loneliness. Loneliness was assessed with a single item. Depression was assessed with the Patient Health Questionnaire-2. Anxiety was assessed with the Generalized Anxiety Disorder-2. Fair/poor general health was assessed by a single item. Linear mixed effects models assessed changes in outcomes after covariate adjustment. RESULTS: The cohort comprised 24,666 individuals (19.2% of older adults in the system). Mean age was 73.7 years (SD 6.4); 54.6% were female, and 11.6% were members of racial and ethnic minority groups. Of these individuals, 1936 (7.8%) were persistently lonely, 1687 (6.8%) were recently lonely, 1551 (6.3%) were previously lonely, and 19,492 (79.0%) were not lonely at either time point. After adjustment for sociodemographic, clinical and social variables, recent loneliness was associated with increases in depression (adjusted odds ratio [aOR] 1.76, 95% confidence interval [CI] 1.41-2.19) and anxiety (aOR 1.67, 95% CI 1.32-2.10). Previous loneliness was associated with decreases in depression (aOR, 0.46, 95% CI 0.36-0.58) and anxiety (aOR 0.69, 95% CI 0.54-0.90). Changes in loneliness were not associated with changes in general health. CONCLUSIONS: Changes in loneliness identified through screening were associated with corresponding changes in depression and anxiety. These findings support the potential value of identifying social risk factors in clinical settings among older adults.

Combination therapy as initial treatment for newly diagnosed hypertension.

BACKGROUND: The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure recommends that clinicians consider the use of multidrug therapy to increase likelihood of achieving blood pressure goal. Little is known about recent patterns of combination antihypertensive therapy use in patients being initiated on hypertension treatment. METHODS: We investigated combination antihypertensive therapy use in newly diagnosed hypertensive patients from the Cardiovascular Research Network Hypertension Registry. Multivariable logistic regression was used to assess the relationship between combination antihypertensive therapy and 12-month blood pressure control. RESULTS: Between 2002 and 2007, a total of 161,585 patients met criteria for incident hypertension and were initiated on treatment. During the study period, an increasing proportion of patients were treated initially with combination rather than with single-agent therapy (20.7% in 2002 compared with 35.8% in 2007, P < .001). This increase in combination therapy use was more pronounced in patients with stage 2 hypertension, whose combination therapy use increased from 21.6% in 2002 to 44.5% in 2007. Nearly 90% of initial combination therapy was accounted for by 2 combinations, a thiazide and a potassium-sparing diuretic (47.6%) and a thiazide and an angiotensin-converting enzyme inhibitor (41.4%). After controlling for relevant clinical factors, including subsequent intensification of treatment and medication adherence, combination therapy was associated with increased odds of blood pressure control at 12 months (odds ratio compared with single-drug initial therapy 1.20; 95% CI 1.15-1.24, P < .001). CONCLUSIONS: Initial treatment of hypertension with combination therapy is increasingly common and is associated with better long-term blood pressure control.

Identifying optimal risk windows for self-controlled case series studies of vaccine safety.

In vaccine safety studies, subjects are considered at increased risk for adverse events for a period of time after vaccination known as risk window. To our knowledge, risk windows for vaccine safety studies have tended to be pre-defined and not to use information from the current study. Inaccurate specification of the risk window can result in either including the true control period in the risk window or including some of the risk window in the control period, which can introduce bias. We propose a data-based approach for identifying the optimal risk windows for self-controlled case series studies of vaccine safety. The approach involves fitting conditional Poisson regression models to obtain incidence rate ratio estimates for different risk window lengths. For a specified risk window length (L), the average time at risk, T(L), is calculated. When the specified risk window is shorter than the true, the incidence rate ratio decreases with 1/T(L) increasing but there is no explicit relationship. When the specified risk window is longer than the true, the incidence rate ratio increases linearly with 1/T(L) increasing. Theoretically, the risk window with the maximum incidence ratio is the optimal risk window. Because of sparse data problem, we recommend using both the maximum incidence rate ratio and the linear relationship when the specified risk window is longer than the true to identify the optimal risk windows. Both simulation studies and vaccine safety data applications show that our proposed approach is effective in identifying medium and long-risk windows.

Association of Exercise Behavior with Overall Survival in Stage I-IIIA Lung Cancer.

Rationale: Exercise assessments may help predict outcomes for patients with diagnosed lung cancer. Objectives: We examined the relationship between prediagnosis exercise behavior and clinical outcomes among patients with stage I-IIIA lung cancer. Methods: In a retrospective cohort study of patients with stage I-IIIA lung cancer at Kaiser Permanente Colorado who had at least one Exercise Vital Sign assessment-a questionnaire tool to help promote exercise in chronic disease management-within the year before diagnosis, we defined exercise behavior as active (any min/wk of moderate-to-vigorous-intensity physical activity) or inactive (no moderate-to-vigorous physical activity). The outcomes were 1) overall survival (OS) and 2) acute healthcare use (AHCU). We used the Kaplan-Meier method, Cox proportional hazard model, and negative binomial regression model to analyze the effects of exercise on outcomes, adjusting for demographic, socioeconomic, clinical, and lung-cancer characteristics. Results: Among 552 patients with lung cancer, 230 (42%) were identified as physically active before their diagnosis of cancer. There was no significant difference in the stage distribution between active and inactive patients. The median survival times were 2.4 years for the active group and 1.8 years for inactive patients (P < 0.001). The mean rates (standard deviations) of AHCU were 1.09 (1.55) and 2.31 (5.61) per person-year for active and inactive groups, respectively (P < 0.01). Active exercise, compared with inactivity, was associated with better OS (hazard ratio, 0.52 [0.39-0.69]) and lower AHCU (rate ratio, 0.63 [0.49-0.80]) in unadjusted analyses; in adjusted analyses, active exercise was associated with better OS (hazard ratio, 0.62 [0.45-0.86]), but AHCU was not lower by a statistically significant amount (rate ratio, 0.82 [0.65-1.04]). Conclusions: Prediagnosis active exercise was associated with better OS after diagnosis of stage I-IIIA lung cancer. Exercise assessments may help predict outcomes, risk-stratify patients for curative-intent therapy, and identify those who would benefit from increased physical activity and exercise.

Factors Associated With Opting Out of Automated Text and Telephone Messages Among Adult Members of an Integrated Health Care System.

Importance: Health care systems deliver automated text or telephone messages to remind patients of appointments and to provide health information. Patients who receive multiple messages may demonstrate message fatigue by opting out of future messages. Objective: To assess whether the volume of automated text or interactive voice response (IVR) telephone messages is associated with the likelihood of patients requesting to opt out of future messages. Design, Setting, and Participants: This retrospective cohort study was conducted at Kaiser Permanente Colorado (KPCO), an integrated health care system. All adult members who received 1 or more automated text or IVR message between October 1, 2018, and September 30, 2019, were included. Exposures: Receipt of automated text or IVR messages. Main Outcomes and Measures: Message volume and opt-out rates obtained from messaging systems over 1 year. Results: Of the 428 242 adults included in this study, 59.7% were women, and 66.5% were White; the mean (SD) age was 52.3 (17.7) years. During the study period, 84.1% received 1 or more text messages (median, 4 messages; interquartile range, 2-8 messages) and 67.8% received 1 or more IVR messages (median, 3 messages; interquartile range, 1-6 messages). A total of 8929 individuals (2.5%) opted out of text messages, and 4392 (1.5%) opted out of IVR messages. In multivariable analyses, individuals who received 10 to 19.9 or 20 or more text messages per year had higher opt-out rates for text messages compared with those who received fewer than 2 messages per year (adjusted odds ratio [aOR]: 10-19.9 vs <2 messages, 1.27 [95% CI, 1.17-1.38]; ≥20 vs <2 messages, 3.58 [95% CI, 3.28-3.91]), whereas opt-out rates increased progressively in association with IVR message volume, with the highest rates among individuals who received 10.0 to 19.9 messages (aOR, 11.11; 95% CI, 9.43-13.08) or 20.0 messages or more (aOR, 49.84; 95% CI, 42.33-58.70). Individuals opting out of text messages were more likely to opt out of IVR messages (aOR, 4.07; 95% CI, 3.65-4.55), and those opting out of IVR messages were more likely to opt out of text messages (aOR, 5.92; 95% CI, 5.29-6.61). Conclusions and Relevance: In this cohort study among adult members of an integrated health care system, requests to discontinue messages were associated with greater message volume. These findings suggest that, to preserve the benefits of automated outreach, health care systems should use these messages judiciously to reduce message fatigue.