A Mixed-Methods Assessment of Coronavirus Disease of 2019-Era Telehealth Acute Care Visits in the Medical Home.

OBJECTIVES: To compare acute care virtual visits with in-person visits with respect to equity of access, markers of quality and safety, and parent and provider experience, before and during the coronavirus disease 2019 pandemic. STUDY DESIGN: We compared patient demographics, antimicrobial prescribing rates, emergency department (ED) use, and patient-experience scores for virtual visits and in-person care at 2 academic pediatric primary care practices using χ2 testing and interrupted time series analyses. Parent and provider focus groups explored themes related to virtual visit experience and acceptability. RESULTS: We compared virtual acute care visits conducted in March 2020-February 2021 (n = 8868) with in-person acute care visits conducted in February 2019-March 2020 (n = 24 120) and March 2020-February 2021 (n = 6054). There were small differences in patient race/ethnicity across the different cohorts (P < .01). Virtual visits were associated with a 9.6% (-11.5%, -7.8%, P < .001) decrease in all antibiotic prescribing and a 13.2% (-22.1%, -4.4%, P < .01) decrease in antibiotic prescribing for acute respiratory tract infections. Unanticipated visits to the ED did not significantly differ among visit types. Patient experience scores were significantly greater (P < .05) for virtual acute care in overall rating of care and likelihood to recommend. Focus group themes included safety, distractibility, convenience, treatment, and technology. Providers were broadly accepting of virtual care while parental views were more mixed. CONCLUSIONS: Telehealth acute care visits may not have negative effects on quality and safety, as measured by antimicrobial prescribing and unanticipated ED visit rates. Efforts to increase parental acceptance and avoid creating disparities in access to virtual care will be essential to continued success of telehealth acute care visits.

Development and validation of the Composite Asthma Severity Index--an outcome measure for use in children and adolescents.

BACKGROUND: Asthma severity is reflected in many aspects of the disease, including impairment and future risks, particularly for exacerbations. According to the Expert Panel Report 3: Guidelines for the Diagnosis and Management of Asthma, however, to assess more comprehensively the severity of asthma the level of current treatment needed to maintain a level of control should be included. OBJECTIVE: Development and validation of a new instrument, the Composite Asthma Severity Index (CASI), which can quantify disease severity by taking into account impairment, risk, and the amount of medication needed to maintain control. At present, there is no instrument available to measure and assess the multidimensional nature of asthma. METHODS: Twenty-six established asthma investigators, who are part of the National Institutes of Health-supported Inner City Asthma Consortium, participated in a modified Delphi consensus process to identify and weight the dimensions of asthma. Factor analysis was performed to identify independent domains of asthma by using the Asthma Control Evaluation trial. CASI was validated by using the Inner City Anti-IgE Therapy for Asthma trial. RESULTS: CASI scores include 5 domains: day symptoms and albuterol use, night symptoms and albuterol use, controller treatment, lung function measures, and exacerbations. At Asthma Control Evaluation trial enrollment, CASI ranged from 0 to 17, with a mean of 6.2. CASI was stable, with minimal change in variance after 1 year of treatment. In external validation, CASI detected a 32% larger improvement than did symptoms alone. CONCLUSION: CASI retained its discriminatory ability even with low levels of symptoms reported after months of guidelines-directed care. Thus, CASI has the ability to determine the level of asthma severity and provide a composite clinical characterization of asthma.

Relationships among environmental exposures, cord blood cytokine responses, allergy, and wheeze at 1 year of age in an inner-city birth cohort (Urban Environment and Childhood Asthma study).

BACKGROUND: The Urban Environment and Childhood Asthma study was established to investigate the immunologic and environmental causes of asthma in inner-city children. OBJECTIVE: We sought to evaluate potential atopic outcomes in the first 12 months and their relationships to environmental exposures and immune development. METHODS: A birth cohort of 560 children with at least 1 parent with allergy or asthma was established in Baltimore, Boston, New York, and St Louis. Wheezing is assessed every 3 months, allergen-specific IgE yearly, and mononuclear cell cytokine responses at birth and yearly; environmental assessments include dust allergen and endotoxin, maternal stress, and indoor nicotine and nitrogen dioxide levels. RESULTS: Key outcomes in the first year include wheeze in 49%, 2 or more episodes of wheeze in 23%, eczema in 30%, and detectable IgE to milk, egg, and/or peanut in 32% and to cockroach in 4%. Household dust revealed levels of greater than 2 µg/g to cockroach in 40%, mite in 19%, cat in 25%, and mouse in 29%, and 66% of homes housed at least 1 smoker. Positive associations were detected between multiple wheeze and cotinine levels, maternal stress, and maternal depression, whereas cytokine responses to a variety of innate, adaptive, and mitogenic stimuli were inversely related to eczema. CONCLUSIONS: This high-risk cohort of inner-city infants is exhibiting high rates of wheeze, eczema, and allergic sensitization. Low cytokine responses at birth might be a risk factor for eczema, whereas a variety of adverse environmental exposures contribute to the risk of wheezing in infancy. These findings provide evidence of specificity in the interactions between immune development, environmental exposures, and the development of early features that might predict future asthma.

Improving delivery of preventative care services using population management strategies.

BACKGROUND: Consistent and timely delivery of comprehensive preventative care services is a challenge, particularly in underserved patient populations. Previous quality improvement (QI) research has focused on the development of bundled measures of preventative services delivery, but these bundles have not been studied on a population level. We aimed to improve preventative care service delivery on a clinic population level through the use of a bundled measure that includes immunisations, lead screening and use of screening tools among underserved patients under 2 years old. METHODS: A QI study was conducted at a community-based academic primary care clinic. A population-level bundled measure was adapted from an existing tool. We used plan-do-study-act cycles to optimise results and tracked bundle outcome rates using a p-type statistical process control chart. Interventions included (1) staff education on measure components, (2) introduction of exam room-based phlebotomy to address lead screening completion rates and (3) population management strategies, including development of a patient registry and use of reminders and visit tracking to increase attendance at well-child visits. RESULTS: The percent of bundle completion by 14 months of age increased from a baseline of 58%-77% following implementation of the QI initiatives. A mean shift was identified after the population manager began proactive targeted outreach for the 12-month visit. CONCLUSION: Targeted systems for outreach aimed at bringing patients into the clinic and patient-centred strategies for visit completion are effective at ensuring timely delivery of comprehensive preventative care to an underserved paediatric population.

Machine learning approaches to predicting no-shows in pediatric medical appointment.

Patients' no-shows, scheduled but unattended medical appointments, have a direct negative impact on patients' health, due to discontinuity of treatment and late presentation to care. They also lead to inefficient use of medical resources in hospitals and clinics. The ability to predict a likely no-show in advance could enable the design and implementation of interventions to reduce the risk of it happening, thus improving patients' care and clinical resource allocation. In this study, we develop a new interpretable deep learning-based approach for predicting the risk of no-shows at the time when a medical appointment is first scheduled. The retrospective study was conducted in an academic pediatric teaching hospital with a 20% no-show rate. Our approach tackles several challenges in the design of a predictive model by (1) adopting a data imputation method for patients with missing information in their records (77% of the population), (2) exploiting local weather information to improve predictive accuracy, and (3) developing an interpretable approach that explains how a prediction is made for each individual patient. Our proposed neural network-based and logistic regression-based methods outperformed persistence baselines. In an unobserved set of patients, our method correctly identified 83% of no-shows at the time of scheduling and led to a false alert rate less than 17%. Our method is capable of producing meaningful predictions even when some information in a patient's records is missing. We find that patients' past no-show record is the strongest predictor. Finally, we discuss several potential interventions to reduce no-shows, such as scheduling appointments of high-risk patients at off-peak times, which can serve as starting point for further studies on no-show interventions.

Randomized Controlled Trial of Health Coaching for Parents of Children With Medical Complexity.

BACKGROUND AND OBJECTIVE: Parents of children with medical complexity (CMCs) struggle with managing their children's care needs. Health coaching provides patients with tools to take ownership of their care. We sought to assess the impact of health coaching for parents of CMCs on activation, child's health related quality of life (HRQL), and utilization. METHODS: Parents of CMCs receiving primary care at two academic primary care practices were randomized to receive 3-6 months of coaching from a trained health coach (n = 54) or usual care (n = 71). Parents were surveyed on activation, HRQL, and experience of care at baseline and the end of the study period. Markers of utilization (missed visit rates, ED, and admissions) were measured for 12 months prior to and after enrollment. Parametric, nonparametric, and Poisson regression were used to assess baseline differences in characteristics and average pre-post differences between the groups. RESULTS: At baseline, activation and rating of patient experience were high in both groups, while HRQL was one standard deviation below the population-based mean. Both groups had slight increases in HRQL and P-PAM without significant differences between groups. ED visits declined significantly more in the intervention group than the usual care group (0.68 visits, (-0.03, -1.32) P = .04); there were nonsignificant differences in changes in other utilization metrics. CONCLUSION: Health coaching was associated with a decrease in ED use but not in changes in activation, patient experience, or HRQL. More inquiry is needed to understand whether health coaching is an effective modality for improving care for children with medical complexity.

Pediatric Primary Care-Based Social Needs Services and Health Care Utilization.

OBJECTIVE: To evaluate the relationship between use of primary care-based social needs services and subsequent utilization of ambulatory, emergency, and inpatient services. METHODS: This retrospective 2012 to 2015 cohort study uses electronic medical record data from an academic pediatric primary care practice that screens universally for social needs and delivers services via in-house social work staff. Logistic regression (N = 7300) examines how patient characteristics relate to practice-based social service use. Negative binomial models with inverse probability of treatment weights (N = 4893) estimate adjusted incidence rate ratios for ambulatory, emergency, and inpatient service use among those who used social services compared to those who did not. RESULTS: Forty-five percent of patients used primary care-based social needs services. This use was significantly greater among those with disabling or complex medical conditions than those without (adjusted odds ratio and 95% confidence interval (CI) of 9.81 [7.39-13.01] and 2.76 [2.44-3.13], respectively); those from low-income versus high-income backgrounds (1.40 [1.21-1.61]); and Blacks and Latinos than Whites (1.33 [1.09-1.62] and 1.29 [1.05-1.59], respectively). Patients who used social services subsequently utilized ambulatory, emergency, and inpatient services at significantly higher rates than those who did not (adjusted incidence rate ratios and 95% CI of 1.54 [1.45-1.63], 1.50 [1.36-1.65], and 3.23 [2.31-4.51], respectively). CONCLUSIONS: Primary care-based social needs service use was associated with increased utilization of ambulatory services without reductions in emergency or inpatient admissions. This pattern suggests increased health care needs or access and could have payment model-dependent financial implications for practices.

Moving the needle on primary care burnout: Using a driver diagram to accelerate impact.

Prior to the COVID-19 pandemic, studies demonstrated an alarming prevalence of burnout in primary care. In the midst of the pandemic, primary care clinician wellbeing deteriorated and burnout rates increased, yet many organizational efforts to reduce burnout were put on hold due to the urgency of the pandemic. In this article, we present the "Reducing Burnout Driver Diagram" as a tool that clinical leaders and policy makers can use to address and mitigate primary care clinician burnout.

Use of Social Psychology to Improve Adherence to National Bronchiolitis Guidelines.

OBJECTIVES: The American Academy of Pediatrics' bronchiolitis guidelines recommend against albuterol and corticosteroids for treating and chest radiographs (CRs) for diagnosing infants with bronchiolitis. However, high rates of nonadherence have been documented. Our objective was to improve guideline adherence in infants with bronchiolitis. METHODS: This quality improvement study was conducted in 1 urban academic pediatric primary care clinic caring for predominately minority and publicly insured children. We tested provider guideline education, display of guidelines in patient care areas, and monthly e-mails to all providers documenting deviation rates, with individual e-mails to providers who deviated. P-charts and interrupted time series analysis were used to estimate the effect of the intervention. RESULTS: There were 380 children <2 years of age with a diagnosis of bronchiolitis in the 16 nonsummer months preintervention and 417 in the 15 postintervention months. Rates of prescribed and administered albuterol declined from 45.7% in the baseline period to 13.7% in the intervention period and CR use dropped from a mean of 10.1% to 3.4%, both demonstrating special cause variation. Steroid use did not change significantly. In interrupted time series analyses, the intervention was associated with a significant decrease in albuterol use (P < .001) but not in CR or steroid use. Emergency department visits declined slightly but admissions for bronchiolitis were stable. CONCLUSIONS: Traditional quality improvement efforts coupled with social psychology techniques resulted in improved guideline adherence in outpatient bronchiolitis management. Additional study will help identify which techniques are most effective for increasing guideline adherence in cases of low-value care.

A Parenting and Life Skills Intervention for Teen Mothers: A Randomized Controlled Trial.

: media-1vid110.1542/5985300176001PEDS-VA_2018-2303Video Abstract BACKGROUND: Teen mothers often present with depression, social complexity, and inadequate parenting skills. Many have rapid repeat pregnancy, which increases risk for poor outcomes. We conducted a randomized controlled trial of a parenting and life skills intervention for teen mothers aimed at impacting parenting and reproductive outcomes. METHODS: Teen mothers were recruited from a teen-tot clinic with integrated medical care and social services. Participants were randomly assigned 1:1 to receive (1) teen-tot services plus 5 interactive parenting and life skills modules adapted from the Nurturing and Ansell-Casey Life Skills curricula, delivered by a nurse and social worker over the infant's first 15 months or (2) teen-tot services alone. A computerized questionnaire was self-administered at intake, 12, 24, and 36 months. Outcomes included maternal self-esteem, parenting attitudes associated with child maltreatment risk, maternal depression, life skills, and repeat pregnancy over a 36-month follow-up. We used generalized linear mixed modeling and logistic regression to examine intervention effects. RESULTS: Of 152 invited, 140 (92%) participated (intervention = 72; control = 68). At 36 months, maternal self-esteem was higher in the intervention group compared with controls (P = .011), with higher scores on preparedness for mothering role (P = .011), acceptance of infant (P = .008), and expected relationship with infant (P = .029). Repeat pregnancy by 36 months was significantly lower for intervention versus control participants. CONCLUSIONS: A brief parenting and/or life skills intervention paired with medical care for teens and their children has positive effects on maternal self-esteem and repeat pregnancy over 36 months.

Association of Perceived Maternal Stress During the Perinatal Period With Electroencephalography Patterns in 2-Month-Old Infants.

Importance: Variation in child responses to adversity creates a clinical challenge to identify children most resilient or susceptible to later risk for disturbances in cognition and health. Advances in establishing scalable biomarkers can lead to early identification and mechanistic understanding of the association of early adversity with neurodevelopment. Objectives: To examine whether maternal reports of stress are associated with patterns in resting electroencephalography at 2 months of age and whether unique electroencephalographic profiles associated with risk and resiliency factors can be identified. Design, Setting, and Participants: For this cohort study, a population-based sample of 113 mother-infant dyads was recruited from January 1, 2016, to March 1, 2018, during regularly scheduled pediatric visits before infants were 2 months of age from 2 primary care clinics in Boston, Massachusetts, and Los Angeles, California, that predominantly serve families from low-income backgrounds. Data are reported from a single time point, when infants were aged 2 months, of an ongoing cohort study longitudinally following the mother-infant dyads. Exposures: Maternal reported exposure to stressful life events and perceived stress. Main Outcomes and Measures: Spectral power (absolute and relative) in different frequency bands (Δ, θ, low and high α, ß, and γ) from infant resting electroencephalography (EEG) and EEG profiles across frequency bands determined by latent profile analysis. Results: Of 113 enrolled infants, 70 (mean [SD] age, 2.42 [0.37] months; 35 girls [50%]) provided usable EEG data. In multivariable hierarchical linear regressions, maternal perceived stress was significantly and negatively associated with absolute ß (ß = -0.007; 95% CI, -0.01 to -0.001; semipartial r = -0.25) and γ power (ß = -0.008; 95% CI, -0.01 to -0.002; semipartial r = -0.28). Maternal educational level was significantly and positively associated with power in high α, ß, and γ bands after adjusting for covariates (high school: γ: ß = 0.108; 95% CI, 0.014-0.203; semipartial r = -0.236; associate's degree or higher: high α: ß = 0.133; 95% CI, 0.018-0.248; semipartial r = 0.241; ß: ß = 0.167; 95% CI, 0.055-0.279; semipartial r = 0.309; and γ: ß = 0.183; 95% CI, 0.066-0.299; semipartial r = 0.323). Latent profile analysis identified 2 unique profiles for absolute and relative power. Maternal perceived stress (ß = 0.13; 95% CI, 0.01-0.25; adjusted odds ratio [AOR], 1.14; 95% CI, 1.01-1.28) and maternal educational level (high school: ß = 3.00; 95% CI, 0.35-5.65; AOR, 20.09; 95% CI, 1.42-283.16; associate's degree or higher: ß = 4.12; 95% CI, 1.45-6.79; AOR, 61.56; 95% CI, 4.28-885.01) were each associated with unique profile membership. Conclusions and Relevance: These findings suggest that unique contributions of caregiver stress and maternal educational level on infant neurodevelopment are detectable at 2 months; EEG might be a promising tool to identify infants most susceptible to parental stress and to reveal mechanisms by which neurodevelopment is associated with adversity. Additional studies validating subgroups across larger cohorts with different stressors and at different ages are required before use at the individual level in clinical settings.

Improving rates of ferrous sulfate prescription for suspected iron deficiency anaemia in infants.

BACKGROUND: Iron deficiency anaemia (IDA) in infancy is prevalent and associated with impaired neurodevelopment; however, studies suggest that treatment and follow-up rates are poor. OBJECTIVES: To improve the rate of ferrous sulfate prescription for suspected IDA among infants aged 8-13 months to 75% or greater within 24 months. METHODS: We implemented a multidisciplinary process improvement effort aimed at standardising treatment for suspected IDA at two academic paediatric primary care clinics. We developed a clinical pathway with screening and treatment recommendations, followed by multiple plan-do-study-act cycles including provider education, targeted reminders when ferrous sulfate was not prescribed and development of standardised procedures for responding to abnormal lab values. We tracked prescription and screening rates using statistical process control charts. In post hoc analyses, we examined rates of haemoglobin (Hgb) recheck and normalisation for the preintervention versus postintervention groups. RESULTS: The prescription rate for suspected IDA increased from 41% to 78% following implementation of the intervention. Common reasons for treatment failure included prescription of a multivitamin instead of ferrous sulfate, and Hgb not flagged as low by the electronic medical record. Screening rates remained stable at 89%. Forty-one per cent of patients with anaemia in the preintervention group had their Hgb rechecked within 6 months, compared with 56% in the postintervention group (p<0.001). Furthermore, 30% of patients with anaemia in the postintervention group had normalised their Hgb by 6 months, compared with 20% in the preintervention group (p<0.05). CONCLUSIONS: A multipronged interdisciplinary quality improvement intervention enabled: (1) development of standardised practices for treating suspected IDA among infants aged 8-13 months, (2) improvement of prescription rates and (3) maintenance of high screening rates. Rates of Hgb recheck and normalisation also increased in the intervention period.​.

Associations Between Patient-Reported Outcome Measures of Asthma Control and Psychosocial Symptoms.

There is growing emphasis on using patient-reported outcome measures to enhance clinical practice. This study was a retrospective review of scores on the Childhood Asthma Control Test (C-ACT) and the Pediatric Symptom Checklist-17 (PSC-17) at a pediatric primary care center in Boston, Massachusetts. A total of 218 patients were selected at random using billing codes for well-child (WC) care and asthma, excluding complex medical conditions. Cutoff scores were used to identify uncontrolled asthma (C-ACT ⩽19) and clinically significant psychosocial symptoms (+PSC-17). Multiple logistic regression was used to measure associations between C-ACT ⩽19 and +PSC-17, adjusting for covariates. In multivariable analysis, C-ACT ⩽19 at WC visits was associated with +PSC-17 at WC visits (adjusted odds ratio = 3.2 [95% confidence interval = 1.3-8.6]). C-ACT ⩽19 at non-WC visits was also associated with +PSC-17 at WC visits (adjusted odds ratio = 3.1 [95% confidence interval = 1.2-8.9]). Patient-reported outcome measures of asthma control and psychosocial symptoms were positively correlated in this sample.

Utilization of Patient Navigators in an Urban Academic Pediatric Primary Care Practice.

Navigating health care systems can be a challenge for families. A retrospective descriptive cohort analysis was conducted assessing referrals to patient navigators (PNs) in one urban academic pediatric primary care practice. PNs tracked referral processes and a subset of PN referrals was assessed for markers of successful referrals. The most common reasons for referral were assistance overcoming barriers to care (46%), developmental concerns (38%), and adherence/care coordination concerns (14%). Significant predictors of referral were younger age, medical complexity, public insurance, male sex, and higher rates of no-show to visits in primary or subspecialist care. The majority of referrals were resolved. The referrals for process-oriented needs were significantly more successful than those for other concerns. PNs were more effective for discrete process tasks than for those that required behavior change by patients or families. Future directions include analysis of cost effectiveness of the PN program and analysis of parent and primary care provider experience.

Ensuring Timely Connection to Early Intervention for Young Children With Developmental Delays.

BACKGROUND AND OBJECTIVES: Timely provision of developmental services can improve outcomes for children 0 to 3 years old with developmental delays. Early Intervention (EI) provides free developmental services to children under age 3 years; however, data suggests that many children referred to EI never connect to the program. We sought to ensure that 70% of patients referred to EI from an academic primary care clinic serving a low-income population were evaluated within 120 days of referral. METHODS: Recognizing that our baseline system of EI referrals had multiple routes to referral without an ability to track referral outcome, we implemented a multifaceted referral process with (1) a centralized electronic referral system used by providers, (2) patient navigators responsible for processing all EI referrals submitted by providers, and (3) a tracking system postreferral to facilitate identification of patients failing to connect with EI. RESULTS: The percentage of patients evaluated by EI within 120 days increased from a baseline median of 50% to a median of 72% after implementation of the systems (N = 309). After implementation, the centralized referral system was used a median of 90% of the time. Tracking of referral outcomes revealed decreases in families refusing evaluations and improvements in exchange of information with EI. CONCLUSIONS: Rates of connection to EI improved substantially when referrals were centralized in the clinic and patient navigators were responsible for tracking referral outcomes. Knowledge of EI intake processes and relationships between the practice and the EI site are essential to ensure successful connections.

Screening for Social Determinants of Health Among Children and Families Living in Poverty: A Guide for Clinicians.

Approximately 20% of all children in the United States live in poverty, which exists in rural, urban, and suburban areas. Thus, all child health clinicians need to be familiar with the effects of poverty on health and to understand associated, preventable, and modifiable social factors that impact health. Social determinants of health are identifiable root causes of medical problems. For children living in poverty, social determinants of health for which clinicians may play a role include the following: child maltreatment, child care and education, family financial support, physical environment, family social support, intimate partner violence, maternal depression and family mental illness, household substance abuse, firearm exposure, and parental health literacy. Children, particularly those living in poverty, exposed to adverse childhood experiences are susceptible to toxic stress and a variety of child and adult health problems, including developmental delay, asthma and heart disease. Despite the detrimental effects of social determinants on health, few child health clinicians routinely address the unmet social and psychosocial factors impacting children and their families during routine primary care visits. Clinicians need tools to screen for social determinants of health and to be familiar with available local and national resources to address these issues. These guidelines provide an overview of social determinants of health impacting children living in poverty and provide clinicians with practical screening tools and resources.

Redesigning Health Care Practices to Address Childhood Poverty.

Child poverty in the United States is widespread and has serious negative effects on the health and well-being of children throughout their life course. Child health providers are considering ways to redesign their practices in order to mitigate the negative effects of poverty on children and support the efforts of families to lift themselves out of poverty. To do so, practices need to adopt effective methods to identify poverty-related social determinants of health and provide effective interventions to address them. Identification of needs can be accomplished with a variety of established screening tools. Interventions may include resource directories, best maintained in collaboration with local/regional public health, community, and/or professional organizations; programs embedded in the practice (eg, Reach Out and Read, Healthy Steps for Young Children, Medical-Legal Partnership, Health Leads); and collaboration with home visiting programs. Changes to health care financing are needed to support the delivery of these enhanced services, and active advocacy by child health providers continues to be important in effecting change. We highlight the ongoing work of the Health Care Delivery Subcommittee of the Academic Pediatric Association Task Force on Child Poverty in defining the ways in which child health care practice can be adapted to improve the approach to addressing child poverty.

High levels of adherence do not prevent accumulation of HIV drug resistance mutations.

OBJECTIVES: To assess the relationship between development of antiretroviral drug resistance and adherence by measured treatment duration, virologic suppression, and the rate of accumulating new drug resistance mutations at different levels of adherence. METHODS: Adherence was measured with unannounced pill counts performed at the participant's usual place of residence in a prospective cohort of HIV-positive urban poor individuals. Two genotypic resistance tests separated by 6 months (G1 and G2) were obtained in individuals on a stable regimen and with detectable viremia (> 50 copies/ml). The primary resistance outcome was the number of new HIV antiretroviral drug resistance mutations occurring over the 6 months between G1 and G2. RESULTS: High levels of adherence were closely associated with greater time on treatment (P < 0.0001) and viral suppression (P < 0.0001) in 148 individuals. In a subset of 57 patients with a plasma viral load > 50 copies/ml on stable therapy, the accumulation of new drug resistance mutations was positively associated with the duration of prior treatment (P = 0.03) and pill count adherence (P = 0.002). Assuming fully suppressed individuals (< 50 copies/ml) do not develop resistance, it was estimated that 23% of all drug resistance occurs in the top quintile of adherence (92-100%), and over 50% of all drug resistance mutations occur in the top two quintiles of adherence (79-100%). CONCLUSION: Increasing rates of viral suppression at high levels of adherence is balanced by increasing rates of drug resistance among viremic patients. Exceptionally high levels of adherence will not prevent population levels of drug resistance.