вход по аккаунту



код для вставкиСкачать
VA L U E I N H E A LT H 2 0 ( 2 0 1 7 ) A 3 9 9 – A 8 1 1 RM3
Methods for Extracting Treatment Patterns for Renal Cell
Carcinoma (RCC) from Social Media (SM) Forums Using Natural
Language Processing (NLP) and Machine Learning (ML)
Merinopoulou E1, Ramagopalan S2, Malcolm B2, Cox A1
Squibb, Uxbridge, UK
1Evidera, London, UK, 2Bristol-Myers
Objectives: Patients are increasingly turning to SM to research their condition
and find support. Patient forums represent a potentially rich source of information on important matters to patients, including treatment. In this study we developed NLP and ML methods to extract and describe treatment histories for RCC
patients. Methods: We collected a corpus of 70,666 posts spanning fifteen years
from 4 popular RCC specific forums. We used ML to identify phrases where patients/
caregivers were describing treatment histories. Classification methods investigated
were: Naïve Bayes, k-Nearest Neighbour and Support Vector Machine. Multiple
direct and derived features were incorporated including bigrams and trigrams.
The selected phrases were tagged with terms from the Unified Medical Language
Thesaurus. Patterns of RCC therapies were extracted manually for a random sample of patients. Results: The ML algorithm selected phrases where patients/caregivers recounted treatment histories with 85% overall accuracy (precision: 94%;
recall: 82%; F-rate: 88%). After sub-setting, the corpus consisted of 13,740 phrases
from 2,384 patients. 50 patients were then randomly selected for manual review,
and compilation of treatment histories. Posting dates ranged from 2006-2016 and
treatment sequence could be determined for 48 patients (96%).The most common
first line therapy was sunitinib (58 %) followed by sorafenib and interleukin (both
13%).22 patients (46%) reported information on 2nd line; most common treatments were: pazopanib, everolimus , sorafenib (all 18%). Most common 3rd line
was everolimus (3/7 patients). Alignment of these findings was seen with published
data. Conclusions: This preliminary work showed that extracting treatment
information from patient forums is challenging but technically feasible. Future
work will focus on improving the accuracy of the ML algorithms, and extending
automation to the assembly of treatment histories for all patients. If successful,
SM can be used to describe real-world treatment sequences that can be especially
useful where data options are limited.
Non-Proportional Hazards in Network Meta-Analysis: Efficient
Strategies for Model Building and Analysis
Gsteiger S1, Windisch R1, Bryden P1, Wiksten A2
Roche Ltd, Basel, Switzerland, 2StatFinn & EPID Research, Espoo, Finland
1F. Hoffmann-La
Objectives: Cancer immunotherapies often show delayed onset of efficacy and
long-term survival benefit compared with chemotherapy. This leads to survival
data violating the proportional hazards assumption. Network meta-analysis
(NMA) of such data should acknowledge this. Two suitable approaches are fractional polynomial (FP) models and piece-wise constant (PWC) models. FP models
can be difficult to fit in practice, and there is a need for efficient model selection strategies. Methods: We re-formulated the FP and PWC NMA models using
ANOVA like parameterization. With this approach, both models are expressed as
generalized linear models with time-changing covariates. We then performed a
case study using our in-house cancer immunotherapy programs. The evidence base
involved 18 studies, some of which linked to the network via Matching Adjusted
Indirect Comparison. First, we fitted fixed effects NMAs in a frequentist framework.
Second, we fitted fixed and random effects versions of the best performing model
in a Bayesian framework. Results: The frequentist fits were extremely fast and
allowed exploration of FPs of different orders and PWCs with different cut-points
very rapidly. The first step involved 6 FP and 8 PWC models, and fitting took less
than a second. Second order FPs had lowest Akaike Information Criterion (AIC) but
suffered from over-fitting. The PWC models with two cut-points were second in
terms of AIC and performed best overall. We then fitted the best PWC model also
in a Bayesian framework, where each model fit took several minutes to converge
and to provide sufficient effective sample size. Conclusions: NMA models with
time-varying hazard ratios can be explored efficiently with a stepwise approach.
A frequentist fixed effects framework and special parameterization enables rapid
exploration of different models. The best model can then be assessed further in
a Bayesian random effects framework to appropriately capture uncertainty and
inform decision making.
Exploring Factors Explaining Treatment Acceptance in Patients
Suffering from a Chronic Disease
Wiederkehr S1, de Bock E2, Chekroun M3, Arnould B4
1Mapi, Patient Centered Sciences, Lyon, France, 2Mapi, Patient-Centered Outcomes, Lyon, France,
3Carenity, Paris, France, 4Mapi, Patient Centered Outcomes, Lyon, France
Objectives: Patients with chronic diseases are generally required to take longterm treatments. However lack of adherence is very common and represents major
barriers to treatment efficiency. Measuring patient acceptance of their medication
help understand and predict patients’ medication-taking behavior. The objectives
of this study are to evaluate the level of acceptance to medication in chronic disease patients in real life; to explore its determinants and identify issues to define
priorities for action. Methods: Observational, cross-sectional study conducted
in Europe using Carenity Online Community. Adult patients with a chronic disease were invited to complete an online questionnaire including a validated
patient reported outcome measure: the 25-item ACCEptance by the Patients of
their Treatment (ACCEPT©). It includes one general acceptance dimension and
five multi-item treatment-attribute specific dimensions scored from 0-100 (lowest
to highest acceptance). Multivariate linear regression was conducted to explore
factors related to Acceptance/General. Results: 3,011 patients were included.
Acceptance/General score was low in average (46.7 ± 33.0) and heterogeneous
across diseases. Acceptance/General score was highly correlated to Acceptance/
Effectiveness (R = 0.61, p < 0.001). The final multivariate linear model explained
a large part of the variation of Acceptance/General levels (N = 3,968; R² = 0.43);
and showed that Acceptance/Effectiveness, Side Effects, Long-Term Treatment (p < 0.001) were significantly and strongly related to it. Socio-demographical and clinical
parameters were also found to be significant but with lower parameter estimates.
The most frequently reported issues were: having difficulty accepting treatment
for the future (39.9%); experiencing side effects (34.5%) and unpleasant side effects
(33.7%). Conclusions: Treatment acceptance is not satisfactory in chronic disease patients. General acceptance is mainly driven by patients’ perceived treatment
effectiveness, side effects and long term use while socio-demographical and clinical
characteristics have a minor contribution. These findings indicate patients’ priorities and unmet needs; however, they must be confirmed using longitudinal data.
The Impact of Pharmacist-LED Medication Therapy Management on
Medication Adherence in Patients with Type 2 Diabetes Mellitus: A
Randomized Controlled Study
Erku DA, Belachew SA, Tegegn HG, Ayele AA
University of Gondar, Gondar, Ethiopia
Objectives: Poor adherence to antidiabetic medications leads to a higher rate of
hospital admissions and adverse health outcomes in patients with type 2 diabetes mellitus. This study aims to evaluate whether a pharmacist-directed medication therapy management, compared to the usual care, could enhance medication
adherence and reduce hospital admission in patients with type 2 diabetes mellitus. Methods: A prospective randomized controlled study was conducted in
patients with type 2 diabetes mellitus from February 1 to July 30, 2016. Patients in
the control group (n = 65) received the usual care while patients in the intervention
group (n = 62) received a personalized pharmacotherapeutic care plan and diabetes
education (medication therapy management). The two groups were compared by
repeated measure ANOVA at 3 and 6-months with medication adherence (using
Morisky medication adherence scale) and number of hospital admissions as the
main outcome variables. Results: A total of 127 patients were included in the
study. A marked and statistically significant increase in medication adherence from
baseline to 3 and 6 months were noted in the intervention group (increased from
9.2% at baseline to 61% at 6 month) compared with the control group (increased
from 13.2% at baseline (to 30.2% at 6 month; p-value< 0.01). In addition, a lower
number of patients in the intervention group had only one hospital admissions
(11 vs 23) or two hospital admissions (7 vs 14) during the study period, compared to
control group. Conclusions: The findings of this study implied that pharmacistled medication therapy management might improve medication adherence and
reduce number of hospitalizations in patients with type 2 diabetes mellitus. Hence,
policies and guidelines should be in place in order for clinical pharmacists to fully
engage in patient care and improve the medication therapy outcomes.
Factors Associated with an Adherence with Antiepileptic Drugs in
Children Treated in Pediatric Practices in Germany
Kostev K1, Dombrowski S1, Jacob L2
am Main, Germany, 2University of Paris 5, Paris, France
1QuintilesIMS, Frankfurt
Objectives: The goal of this study was to analyze adherence to antiepileptic drugs (AED) in children and adolescents treated in pediatric practices in
Germany. Methods: The present study included patients aged between 2 and 17
years who were diagnosed with epilepsy (ICD 10: G40) and had received at least
two prescriptions of AED between January 2006 and December 2015 in 243 pediatric
practices in Germany. The medication possession ratio (MPR) was used to estimate
adherence, and patients with a MPR greater than 80% were considered adherent. The
impact of patient and drug characteristics on adherence was analyzed using a multivariate logistic regression model. Results: A total of 5,214 patients were included.
Mean age was 10.9 years (SD= 4.9 years). The overall MPR was 88.8% (SD= 34.1%), and
68.9% of patients were considered adherent. Children aged 5 years or younger were
more adherent to AED than those aged between 14 and 17 years (OR= 1.22, 95% CI:
1.07-1.39). Individuals living in western Germany were also found to be more adherent than those living in eastern Germany (OR= 1.71, 95% CI: 1.55-1.88). Asthma as a
comorbidity (OR= 1.59, 95% CI: 1.29-1.96) was positively and ADHD (OR= 0.81, 95% CI:
0.71-0.93) negatively associated with treatment adherence. Finally, no significant
association was found between adherence and the type of AED. Conclusions:
Two-thirds of children and adolescents suffering from epilepsy in Germany were
adherent to AED. Age, place of residence, and comorbidities were significantly associated with adherence.
Predictive Analysis for Identifying Patient Characteristics
Associated with Primary Medication Nonadherence for Lipid
Lowering Therapies
Hill JW1, Rane PB2, Hines DM1, Patel J2, Harrison DJ2, Wade RL1
1QuintilesIMS, Plymouth Meeting, PA, USA, 2Amgen Inc., Thousand Oaks, CA, USA
Objectives: Primary medication nonadherence (PMN) occurs when patients prescribed a medication by a physician fail to fill their prescription. The study objective
was to identify patient characteristics associated with a high risk of PMN for lipid
lowering therapies (LLT). Methods: PMN for LLT was estimated by identifying
patients with new prescription orders for statin and/or ezetimibe prescriptions
in a large U.S. electronic medical records database between 7/1/2013 - 7/31/2015,
Без категории
Размер файла
59 Кб
2017, 021, jval
Пожаловаться на содержимое документа