ABSTRACT
Background: Evidence suggests negative monthly medication adherence trends during the COVID-19 era for patients with HIV (PWH) and multiple chronic conditions. However, it is unknown whether observed trends are associated with changes in outcomes of HIV care before and during the COVID-19 era. Method(s): Adult PWH with type 2 diabetes, hypertension, and/or hypercholesterolemia were identified in a US mid-Atlantic integrated health system. Multivariable population-averaged panel general estimating equations were used to assess the relationship between medication adherence [i.e., accepted dichotomous thresholds for optimal proportion of days covered (PDC)] for four medication groups: antiretrovirals [ART], diabetes medications [DMs], renin-angiotensin antagonists [RASMs], and statins [SMs] and their related clinical endpoints [i.e., viral load (VL;copies/mL), HbA1c, systolic and diastolic blood pressure (SBP, DBP;mmHg), and total cholesterol (TC;mg/dl)] during a 37-month longitudinal observation period [9/2018-9/2021]. Covariates included demographics, number of medication groups, COVID-19 era (starting 3/1/2020), and a COVID-19/PDC interaction term. Result(s): The cohort [n=543] was predominantly 51-64y [59.30%], Black [73.11%], male [69.24%], and privately insured [65.38%]. All patients were prescribed ART with 75.32% co-prescribed SMs;followed by RASMs [42.73%];and DMs [25.60%]. ART PDC>=0.9 was associated with decreased odds of VL>=200 copies/mL [aOR=0.77, 95% CI: 0.63-0.94]. For DMs, RASMs and SMS, PDC>=0.8 was not associated with the clinical endpoints of HbA1c>=7.0% [aOR=0.99, 95% CI: 0.94-1.04], SBP>=130 mmHg [aOR=1.03;95% CI: 0.93-1.14], DBP>=80 mmHg [aOR=1.05, 95% CI: 0.94-1.16] or TC>=200 mg/dl [aOR=1.00, 95% CI: 0.96-1.04], respectively. The COVID-19 era [3/2020 to 9/2021] was associated with increased odds of SBP>=130 [aOR=1.22, 95% CI: 1.01-1.48], but not for DBP>=80 mmHg [aOR=1.05, 95% CI: 0.85-1.28], VL>=200 copies/ mL [aOR=1.01, 95% CI: 0.67-1.52], HbA1c>=7.0% [aOR=0.99, 95% CI: 0.88-1.11], and TC>=200 mg/dl [aOR=0.95, 95% CI: 0.86-1.04]. No interactions between COVID-19 era and PDC on clinical endpoints were observed. Conclusion(s): ART adherence was associated with viral suppression in PWH, but there were no observed associations between DM, RASM, and SM adherence and their respective clinical endpoints. With the exception of a direct relationship between the COVID-19 start date and SBP, the COVID-19 era was not associated with variations in VL, HbA1c, DBP, and TC clinical endpoints.
ABSTRACT
INTRODUCTION: Data on COVID-19 in pregnancy are skewed toward infection at time of labor and delivery, and few studies have controlled for confounding variables. We aimed to explore the sociodemographic and health risk factors for COVID-19 at any point in pregnancy and its impact on maternal outcomes in a diverse cohort during the first year of the pandemic. METHODS: We conducted a retrospective cohort study using data ed from the electronic medical record within Kaiser Permanente Mid-Atlantic States, an integrated health care system, from March 15, 2020, to March 15, 2021. We included women at least 15 years old and pregnant during that timeframe, comparing those who tested positive for COVID-19 to those who did not.We usedmultivariable logistic regression to identify risk factors for COVID-19 infection during pregnancy.We then used propensity score matching to create a comparison group to explore associations between infection and key outcomes. RESULTS: Among 18,285 pregnant members, 1,036 (5.7%) tested positive for COVID-19 during pregnancy, with 26%, 31%, and 43%, respectively, diagnosed in each trimester of pregnancy. Patients with COVID-19 were more likely to be young, Latina, obese, and multiparous;being White or Asian was protective (P<.001). Patients with COVID-19 during pregnancy were more likely to be hospitalized apart from delivery (P=.029). There were no significant differences in fetal demise, cesarean delivery, preterm delivery, gestational diabetes, hypertensive disorders of pregnancy, venous thromboembolism, postpartum readmission, or maternal death between groups. CONCLUSION: Our study was consistent with previously identified disparities in COVID-19 infection. Outcome data were surprisingly reassuring.
ABSTRACT
Background: The natural history of the longer-term effects of SARS-CoV-2 (COVID-19), known as Post-Acute Sequelae of SARS-CoV-2 (PASC), is limited. Disease characterization and definition changed over time and identification via standard diagnosis codes was only recently enacted. We aim to identify a cohort of individuals with, or at-risk for, PASC among Kaiser Permanente Mid-Atlantic States (KPMAS) members, and to identify the clinical conditions of greater burden for those with PASC. Methods: Within our electronic health record system (including internal/external records), we identified adult patients (≥18 years) who had a detectable SARS-CoV-2 RT-PCR result between 1/1/2020-12/31/2020. Non-COVID disease diagnoses/conditions were categorized into specific time intervals based on the first positive SARS-CoV-2 test as the index date (T0), defined as: 1) "prevalent": diagnoses in 4 years prior to T0 and excluded from later consideration;2) "persistent/acute": new disease diagnoses 0-30 days post-T0 and persisted 30-120 days further, and not included as prevalent;3) "incident/late": new disease diagnoses 30-120 days post-T0, not previously identified as prevalent or persistent/acute. Diagnoses were grouped using Clinical Classification Software (CCS) to isolate conditions for PASC. Final CCS distributions were computed relative to the condition counts for each time interval, validated by infectious disease physicians to identify conditions of focus (COF). Results: From the resulting 31,390 patients, we identified the 14 most common COF (Table 1). The most common persistent/acute COF were other lower respiratory disease (4.5%) and respiratory failure (2.7%). Most common incident/late COF (i.e., >2.0% of those testing COVID+) were abdominal pain, gastrointestinal disorders, other nervous system disorders, nonspecific chest pain, dizziness or vertigo, malaise and fatigue, anxiety disorders, mental health disorders, other lower respiratory disease (not previously diagnosed), and cardiac dysrhythmias. No other COF were >2.0% in the persistent or incident time periods. Conclusion: We have identified conditions clinically associated with COVID-19 that persist from infection or present as incident beyond the acute COVID-19 period. This condition list should be utilized in clinical practice when following up with COVID-19 patients. Further research is needed to understand how these conditions compare to people who did not have COVID-19 and to describe their severity, persistence, and resolution.
ABSTRACT
Background: Post-Acute Sequelae of SARS-CoV-2 Infection (PASC) is a novel condition generally defined as new onset or persistence of symptoms related to SARS-CoV-2 beyond convalescence or first 30 days post-diagnosis. PASC has not been well defined by conditions or timeline manifestation. We measured PASC incidence in an integrated health system population (Kaiser Permanente Mid-Atlantic States;KPMAS) and provided supporting evidence for PASC-related conditions of focus (COF) identified from our previous research. Importantly, KPMAS is a closed healthcare system with high ascertainment of COVID-19 among our members, as well as PASC conditions and symptoms. Methods: Using KPMAS electronic health records, we identified adult patients (≥18 years) who had a SARS-CoV-2 RT-PCR test result (detected or undetected) from 1/1/2020 to 12/31/2020. We defined 3 diagnostic time intervals, predicated on the first test date of identified PASC phenotypes. These time intervals were defined as: T1) "Prevalent": 4 years prior to PCR test identifying prevalent conditions;T2) "Acute/Persistent": 0-30 days post-PCR and persisted in 30-120 day follow-up;T3) "Incident/Late": 30-120 days post-PCR identifying incident conditions/symptoms. We enforced mutual exclusivity per patient by removing conditions and symptoms from T2 previously identified in T1 and those from T3 previously identified in T1 or T2. Diagnoses were grouped using Clinical Classification Software (CCS). The PCR-positive patients (cases) were matched to PCR-negative patients (controls) by month of test, age group, race, sex, and medical center. We prioritized 1:3 (case:control) matching, followed by 1:2, then 1:1. Risk ratios with 95% confidence intervals comparing case to control COF were calculated to determine significant COF. Results: Matching successfully resulted in 28,118 cases and 70,293 controls. Demographic differences were negligible and showed no association (Highest Cramer's V: Age-.0511). Overall, risk of COF was 12% greater among cases than controls (Table 1). During T3, risk was significantly higher among cases for the following COF: anosmia, cardiac dysrhythmia, diabetes, genitourinary disorders, malaise, and nonspecific chest pain. Conclusion: We delineated significant COF among those experiencing incident PASC in our KPMAS population. Our findings contribute to the overall evaluation of PASC and provide supporting evidence for an accepted definition. Further understanding the severity and duration of these conditions will be crucial.
ABSTRACT
Background: It is not known if people with HIV (PWH) in the United States (US) have different access to SARS-CoV-2 RT-PCR (COVID-19) testing, or positivity proportions (among those tested), than people without HIV (PWOH). We describe COVID-19 testing and positivity proportions in 6 large geographically and demographically diverse cohorts of PWH and PWOH. Methods: The Corona-Infectious-Virus Epidemiology Team (CIVET) is comprised of five COVID-19 clinical cohorts within a health system (Kaiser Permanente Northern California, Oakland, CA;Kaiser Permanente Mid-Atlantic States, Rockville, MD;University of North Carolina Health, Chapel Hill, NC;Vanderbilt University Medical Center, Nashville, TN;Veterans Aging Cohort Study) and one established classical HIV cohort (MACS/WIHS Combined Cohort Study). Each participating cohort is restricted to individuals who were alive and “in-cohort” in 2020 (definitions of which were operationalized to fit the structure of each cohort). We calculated the percentage of patients in-cohort who were COVID-19 tested, and the proportion COVID-19 positive monthly, by HIV status, from March 1 to August 31, 2020. We report findings from the classical cohort separately because results are based on self-reported information. Results: In the 5 clinical cohorts, PWH ranged from N=2,515 to 31,040, and N=77,019 to 3,710,360 PWOH. Over the 6 month study period, the percentage of PWH who were tested for COVID-19 (13.5%-21.2%) was slightly higher than PWOH (10.8%-14.3%) in each of the cohorts (p-values in each cohort <0.001). However, among those tested, the percentage of patients with positive COVID-19 tests was similar regardless of HIV status (Figure). In the classical cohort that contributed self-reported testing and positive information (PWH N=2,222;PWOH N=1,417), the proportion tested was similar by HIV status (PWH 38.1% vs. PWOH 37.4%), but PWH had a greater positivity proportion (9.0%) compared with PWOH (5.3%, p-value=0.012). Conclusion: Although PWH had higher testing rates compared with PWOH, we did not find evidence of increased positivity among those tested in 5 clinical cohorts with large diverse populations across the US. We will continue to monitor testing, positivity, and COVID-19 related health outcomes in PWH and PWOH using our multiple data sources and leveraging the expertise of established longitudinal cohort studies in the CIVETS collaboration.