Dataset Information

Identifying routine clinical predictors of non-adherence to second-line therapies in type 2 diabetes: A retrospective cohort analysis in a large primary care database.

ABSTRACT:

Aims

To investigate whether combinations of routinely available clinical features can predict which patients are likely to be non-adherent to diabetes medication.

Materials and methods

A total of 67?882 patients with prescription records for their first and second oral glucose-lowering therapies were identified from electronic healthcare records (Clinical Practice Research Datalink). Non-adherence was defined as a medical possession ratio (MPR) ?80%. Potential predictors were examined, including age at diagnosis, sex, body mass index, duration of diabetes, glycated haemoglobin, Charlson index and other recent prescriptions.

Results

Routine clinical features were poor at predicting non-adherence to the first diabetes therapy (c-statistic = 0.601 for all in combined model). Non-adherence to the second drug was better predicted for all combined factors (c-statistic =0.715) but this improvement was predominantly a result of including adherence to the first drug (c-statistic =0.695 for this alone). Patients with an MPR ?80% for their first drug were 3.6 times (95% confidence interval 3.3,3.8) more likely to be non-adherent to their second drug (32% vs. 9%).

Conclusions

Although certain clinical features were associated with poor adherence, their performance for predicting who is likely to be non-adherent, even when combined, was weak. The strongest predictor of adherence to second-line therapy was adherence to the first therapy. Examining previous prescription records could offer a practical way for clinicians to identify potentially non-adherent patients and is an area warranting further research.

SUBMITTER: Shields BM

PROVIDER: S-EPMC6916179 | biostudies-literature | 2020 Jan

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Identifying routine clinical predictors of non-adherence to second-line therapies in type 2 diabetes: A retrospective cohort analysis in a large primary care database.

Shields Beverley M BM Hattersley Andrew T AT Farmer Andrew J AJ

Diabetes, obesity & metabolism 20191007 1

<h4>Aims</h4>To investigate whether combinations of routinely available clinical features can predict which patients are likely to be non-adherent to diabetes medication.<h4>Materials and methods</h4>A total of 67 882 patients with prescription records for their first and second oral glucose-lowering therapies were identified from electronic healthcare records (Clinical Practice Research Datalink). Non-adherence was defined as a medical possession ratio (MPR) ≤80%. Potential predictors were exam ...[more]

PMID: 31468676

Similar Datasets

Project description:BackgroundPatients with heart failure (HF) are the most commonly readmitted group of adult patients in Germany. Most patients with HF are readmitted for noncardiovascular reasons. Understanding the relevance of HF management outside the hospital setting is critical to understanding HF and factors that lead to readmission. Application of machine learning (ML) on data from statutory health insurance (SHI) allows the evaluation of large longitudinal data sets representative of the general population to support clinical decision-making.ObjectiveThis study aims to evaluate the ability of ML methods to predict 1-year all-cause and HF-specific readmission after initial HF-related admission of patients with HF in outpatient SHI data and identify important predictors.MethodsWe identified individuals with HF using outpatient data from 2012 to 2018 from the AOK Baden-Württemberg SHI in Germany. We then trained and applied regression and ML algorithms to predict the first all-cause and HF-specific readmission in the year after the first admission for HF. We fitted a random forest, an elastic net, a stepwise regression, and a logistic regression to predict readmission by using diagnosis codes, drug exposures, demographics (age, sex, nationality, and type of coverage within SHI), degree of rurality for residence, and participation in disease management programs for common chronic conditions (diabetes mellitus type 1 and 2, breast cancer, chronic obstructive pulmonary disease, and coronary heart disease). We then evaluated the predictors of HF readmission according to their importance and direction to predict readmission.ResultsOur final data set consisted of 97,529 individuals with HF, and 78,044 (80%) were readmitted within the observation period. Of the tested modeling approaches, the random forest approach best predicted 1-year all-cause and HF-specific readmission with a C-statistic of 0.68 and 0.69, respectively. Important predictors for 1-year all-cause readmission included prescription of pantoprazole, chronic obstructive pulmonary disease, atherosclerosis, sex, rurality, and participation in disease management programs for type 2 diabetes mellitus and coronary heart disease. Relevant features for HF-specific readmission included a large number of canonical HF comorbidities.ConclusionsWhile many of the predictors we identified were known to be relevant comorbidities for HF, we also uncovered several novel associations. Disease management programs have widely been shown to be effective at managing chronic disease; however, our results indicate that in the short term they may be useful for targeting patients with HF with comorbidity at increased risk of readmission. Our results also show that living in a more rural location increases the risk of readmission. Overall, factors beyond comorbid disease were relevant for risk of HF readmission. This finding may impact how outpatient physicians identify and monitor patients at risk of HF readmission.

Project description:ObjectiveThis study described the demographics, treatment information and identified characteristics associated with virological failure and being lost to follow-up (LTFU) for patients with HIV on first-line and second-line antiretroviral therapy (ART) regimens in a large South African cohort.DesignA quantitative retrospective cohort study using secondary data analysis.SettingSeven Johannesburg inner city facilities.ParticipantsUnique records of 123 002 people with HIV receiving ART at any point in the period 1 April 2004 to 29 February 2020 were included.MeasuresDemographic characteristics, ART status, CD4 count information and retention status were collected and analysed as covariates of outcomes (viral load (VL) and LTFU).ResultsOf the total study patients, 95% (n=1 17 260) were on a first-line regimen and 5% (n=5742) were on a second-line regimen. Almost two-thirds were female (64%, n=79 226). Most patients (60%, n=72 430) were initiated on an efavirenz-based, tenofovir disoproxil fumarate-based and emtricitabine-based regimen (fixed-dose combination). 91% (n=76 737) achieved viral suppression at least once since initiating on ART and 60% (n=57 981) remained in care as at the end of February 2020. Patients from the community health centre and primary healthcare clinics were not only more likely to be virally suppressed but also more likely to be LTFU. Patients on second-line regimens were less likely to reach viral suppression (adjusted OR (aOR)=0.26, CI=0.23 to 0.28) and more likely to be LTFU (aOR=1.21, CI=1.09 to 1.35). Being older (≥25 years) and having a recent CD4 cell count≥100 cells/µL were predictors of viral suppression and retention in patients on ART.ConclusionPatients on first-line regimens had higher VL suppression rates and were more likely to remain in care than those on a second-line regimen. Being younger and having low CD4 cell counts were associated with poor outcomes, suggesting priority groups for ART adherence support.

Project description:Scarce evidence exists about the best treatment for multi-system inflammatory syndrome (MIS-C). We analyzed the effects of steroids, intravenous immunoglobulin (IVIG), and their combination on the probability of discharge over time, the probability of switching to second-line treatment over time, and the persistence of fever 2 days after treatment. We did a retrospective study to investigate the effect of different treatments on children with MIS-C from 1 March 2020 to 1 June 2021. We estimated the time-to-event probability using a Cox model weighted by propensity score to balance the baseline characteristics. Thirty of 132 (22.7%) patients were initially treated with steroids alone, 29/132 (21.9%) with IVIG alone, and 73/132 (55%) with IVIG plus steroids. The probability of early discharge was higher with IVIG than with IVIG plus steroids (hazard ratio [HR] 1.65, 95% CI 1.11-2.45, p = 0.013), but with a higher probability of needing second-line therapy compared to IVIG plus steroids (HR 3.05, 95% CI 1.12-8.25, p = 0.028). Patients on IVIG had a higher likelihood of persistent fever than patients on steroids (odds ratio [OR] 4.23, 95% CI 1.43-13.5, p = 0.011) or on IVIG plus steroids (OR 4.4, 95% CI 2.05-9.82, p < 0.001). No differences were found for this endpoint between steroids or steroids plus IVIG. Conclusions: The benefits of each approach may vary depending on the outcome assessed. IVIG seemed to increase the probability of earlier discharge over time but also of needing second-line treatment over time. Steroids seemed to reduce persistent fever, and combination therapy reduced the need for escalating treatment. What is Known: • Steroids plus intravenous immunoglobulin, compared with intravenous immunoglobulin alone for multi-system inflammatory syndrome (MIS-C) might reduce the need for hemodynamic support and the duration of fever, but the certainty of the evidence is low. What is New: • Intravenous immunoglobulin, steroids, and their combination for MIS-C may have different outcomes. • In this study, intravenous immunoglobulin increased the probability of discharge over time, steroids reduced persistent fever, while combination therapy reduced the need for second-line treatments.

Dataset Information

Identifying routine clinical predictors of non-adherence to second-line therapies in type 2 diabetes: A retrospective cohort analysis in a large primary care database.

Aims

Materials and methods

Results

Conclusions

Publications

Identifying routine clinical predictors of non-adherence to second-line therapies in type 2 diabetes: A retrospective cohort analysis in a large primary care database.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets