Predicting health outcomes in dogs using insurance claims data.
Ontology highlight
ABSTRACT:
SUBMITTER: Debes C
PROVIDER: S-EPMC10240479 | biostudies-literature | 2023 Jun
REPOSITORIES: biostudies-literature
Publications
Predicting health outcomes in dogs using insurance claims data.
Debes Christian C Wowra Johannes J Manzoor Sarosh S Ruple Audrey A
Scientific reports 20230605 1
In this paper we propose a machine learning-based approach to predict a multitude of insurance claim categories related to canine diseases. We introduce several machine learning approaches that are evaluated on a pet insurance dataset consisting of 785,565 dogs from the US and Canada whose insurance claims have been recorded over 17 years. 270,203 dogs with a long insurance tenure were used to train a model while the inference is applicable to all dogs in the dataset. Through this analysis we de ...[more]
Similar Datasets
Project description:This study investigated neurodevelopment and risk factors in children surgically treated for congenital gastrointestinal anomalies (CGIA), excluding those with known high-risk factors such as low birth weight or chromosomal anomalies. Data of children born between 2008 and 2015 who underwent surgical treatment for CGIA were retrieved from the Korean National Health Insurance Database. CGIA included esophageal atresia, duodenal atresia, jejunoileal atresia, anorectal malformations, and congenital megacolon. Neurodevelopmental impairment (NDI) was defined as Korean Ages and Stages Questionnaire scores below the determined cut-off or Korean Developmental Screening Test scores < 2 standard deviations at 3 years of age. Children with CGIA had a significantly higher risk of NDI than controls (6.2% vs. 2.7%, p < 0.001). Growth failure was correlated with NDI. Longer durations of oxygen support (adjusted odds ratio [aOR], 1.037; 95% confidence interval [CI], 1.013-1.063), mechanical ventilation (aOR, 1.053; 95% CI, 1.018-1.089), and number of surgeries (aOR, 1.137; 95% CI, 1.016-1.273) were significantly associated with NDI. These findings emphasize that cautious yet proactive neurodevelopmental monitoring is crucial in affected children, ensuring timely intervention and that excessive concern among families is unnecessary.
| S-EPMC11461865 | biostudies-literature
Project description:BackgroundAlthough the trend of progressing morbidity is widely recognized, there are numerous challenges when studying multimorbidity and patient complexity. For multimorbid or complex patients, prone to fragmented care and high health care use, novel estimation approaches need to be developed.ObjectiveThis study aims to investigate the patient multimorbidity and complexity of Swiss residents aged ≥50 years using clustering methodology in claims data.MethodsWe adopted a clustering methodology based on random forests and used 34 pharmacy-based cost groups as the only input feature for the procedure. To detect clusters, we applied hierarchical density-based spatial clustering of applications with noise. The reasonable hyperparameters were chosen based on various metrics embedded in the algorithms (out-of-bag misclassification error, normalized stress, and cluster persistence) and the clinical relevance of the obtained clusters.ResultsBased on cluster analysis output for 18,732 individuals, we identified an outlier group and 7 clusters: individuals without diseases, patients with only hypertension-related diseases, patients with only mental diseases, complex high-cost high-need patients, slightly complex patients with inexpensive low-severity pharmacy-based cost groups, patients with 1 costly disease, and older high-risk patients.ConclusionsOur study demonstrated that cluster analysis based on pharmacy-based cost group information from claims-based data is feasible and highlights clinically relevant clusters. Such an approach allows expanding the understanding of multimorbidity beyond simple disease counts and can identify the population profiles with increased health care use and costs. This study may foster the development of integrated and coordinated care, which is high on the agenda in policy making, care planning, and delivery.
| S-EPMC9016510 | biostudies-literature
Project description:Type II diabetes mellitus is a global public health challenge, necessitating robust epidemiological investigations. The majority of evidence reports prevalence as estimations of incidence requiring longitudinal cohort studies that are challenging to conduct. However, this has been addressed by the secondary use of existing health insurance claims data. The current study aimed to examine the incidence of type II diabetes mellitus using existing claims and ledger data. The National Health Insurance and medical care system databases were used to extract type II diabetes mellitus (defined as ICD10 codes E11$-14$) claims data over a period of 5 years for individuals over 40 years old living in one city in Japan. Prevalence was calculated, and insured individuals whose data could be tracked over the entire study period were included in the subsequent analyses. Therefore, annual incidence was calculated by estimating differences in prevalence by year. Data analyses were stratified by sex and age group, and a model analysis was conducted to account for these variables. Overall, the prevalence, diabetes medication usage, and insulin usage were 26.3%, 12.1%, and 2.0%, respectively. Annual incidence of type II diabetes mellitus ranged between 1.2% and 4.6%. Both prevalence and incidence tended to be higher in males and peaked around 60-80 years old. The overall annual incidence was estimated at 3.03% (95% CI: 2.21%-3.85%). The annual incidence was not always associated with a low risk, indicating a consistent risk from middle age onward, although the level of risk varied with age. The current study successfully integrated existing claims and ledger data to explore incidence, and this methodology could be applied to a range of injuries and illnesses in the future.
| S-EPMC11446451 | biostudies-literature
Project description:BackgroundIdiopathic pulmonary fibrosis (IPF) is a life-threatening interstitial lung disease (ILD). Characterizing health outcomes of IPF patients is challenging due to disease rarity.ObjectiveThis study aimed to identify the burden of disease in patients newly diagnosed with IPF.MethodsPatients with ≥1 claim with an IPF diagnosis were identified from a United States healthcare insurer's database (2000-2013). Patients with other known causes of ILD or aged <40 years were excluded. Subgroups were compared based on the 2011 change in International Classification of Diseases, 9th Revision (ICD-9) definition of IPF and occurrence of IPF testing. The prevalence and incidence of preselected health conditions of clinical interest were estimated.ResultsMedian age of newly diagnosed patients (n = 7,298) was 62 years (54.0% male). Restricting to patients with IPF diagnostic testing did not substantially affect cohort characteristics, nor did ICD-9 IPF coding change. Mean follow-up was 1.7 years; 16.8% of patients died; and a substantial proportion of patients were censored due to end of health plan enrollment (50.7%) and other causes of ILD (19.6%). The incidence of pulmonary hypertension, lung cancer, and claims-based algorithm proxy for acute respiratory worsening of unknown cause was 22.5, 17.6, and 12.6 per 1,000 person-years, respectively.ConclusionsPatients with IPF had a high disease burden with a variety of health outcomes observed, including a high rate of mortality. Database censoring due to changes in enrollment or other ILD diagnoses limited follow-up. Altering cohort entry definitions, including IPF testing or ICD-9 IPF coding change, had little impact on cohort baseline characteristics.
| S-EPMC7949201 | biostudies-literature
Project description:BackgroundAlthough strengthening coverage has improved cancer care, there are concerns related to medical distortion. Previous studies have only examined whether patients visit a specific hospital, and not the continuum of patients with cancer, resulting in a lack of evidence in South Korea. This study aimed to investigate the patterns in hospital type for cancer care and analyze their association with outcomes.MethodsThe data for this study were obtained from the National Health Insurance Services Sampled Cohort database. This study included patients with four types of cancer (top four cancer incidence in 2020): gastric (3353), colorectal (2915), lung (1351), and thyroid (5158) cancer. The latent class mixed model was used to investigate cancer care patterns, and multiple regression or survival analysis was performed to examine medical cost, length of stay (LOS), and mortality.ResultsThe patterns in each cancer type were classified into two to four classes, namely, mainly visited clinics or hospitals, mainly visited general hospitals, mainly visited tertiary hospitals (MT), and tertiary to general hospitals through trajectory modeling based on the utilization of cancer care. Compared to the MT pattern, other patterns were generally associated with higher cost, LOS, and mortality.ConclusionThe patterns found in this study may be a more realistic way of defining patients with cancer in South Korea compared to previous studies, and its association-related outcomes may be used as a basis to address problems in the healthcare system and prepare alternatives for patients with cancer. Future studies should review cancer care patterns related to other factors such as regional distribution.
| S-EPMC10358188 | biostudies-literature
Project description:BackgroundSince antimicrobial overuse and misuse can have substantial impacts on both public health and fetal well-being, it is essential to gain comprehensive insights into antimicrobial consumption patterns in pregnant women. This study aims to demonstrate antimicrobial utilization in pregnant women.MethodsWe conducted a population-based cohort study using National Health Insurance claims data from January 2009 to December 2020 in South Korea. The target population was pregnancies in women aged 15-45 years who gave birth between 2011 and 2019. The outcome measure was the percentage of antibiotic prescriptions by trimester, subgroup, diagnostic category, and therapeutic category. Antibiotics were defined as J01 in the WHO ATC/DDD classification. To analyze the factors that influenced antibiotic prescriptions, we conducted multivariate logistic regression analysis.ResultsAntibiotics were prescribed at least once in 1,808,588 (50%) of the 3,614,478 pregnancies. The proportion of deliveries with exposure to antibiotic therapy during pregnancy increased from 48% in 2011 to 54.8% in 2020. The prescription rate of antibiotics was highest in mothers younger than 25 years old, and it was lowest in participants aged 30-34 years. Also, antibiotic use was highest in the first trimester (30%) and lowest in the second trimester (18.2%). The most commonly used class was J01D (other beta-lactam antibacterials), which includes cephalosporins, and it accounted for 57% of the prescribed antibiotics. An increased probability of being prescribed antibiotics was associated with those younger than 25 years old, insurance (Medical Aid), hospitalization experience, an increase in physician visits, and comorbidities. For comorbidities, the OR was higher for respiratory tract infection (RTI), sexually transmitted infection, and urinary tract infection compared to those without disease, whereas it was decreased for diabetes and epilepsy.ConclusionThe prescribing rate of antibiotics to pregnant women in South Korea has remained stably high. Also, an increase in the use of broad-spectrum beta-lactam penicillin was noted. It is necessary to monitor antibiotics in pregnant women generally in accordance with recommendations.
| S-EPMC10451209 | biostudies-literature
Project description:BackgroundDecomposing health care spending by disease, type of care, age, and sex can lead to a better understanding of the drivers of health care spending. But the lack of diagnostic coding in outpatient care often precludes a decomposition by disease. Yet, health insurance claims data hold a variety of diagnostic clues that may be used to identify diseases.MethodsIn this study, we decompose total outpatient care spending in Switzerland by age, sex, service type, and 42 exhaustive and mutually exclusive diseases according to the Global Burden of Disease classification. Using data of a large health insurance provider, we identify diseases based on diagnostic clues. These clues include type of medication, inpatient treatment, physician specialization, and disease specific outpatient treatments and examinations. We determine disease-specific spending by direct (clues-based) and indirect (regression-based) spending assignment.ResultsOur results suggest a high precision of disease identification for many diseases. Overall, 81% of outpatient spending can be assigned to diseases, mostly based on indirect assignment using regression. Outpatient spending is highest for musculoskeletal disorders (19.2%), followed by mental and substance use disorders (12.0%), sense organ diseases (8.7%) and cardiovascular diseases (8.6%). Neoplasms account for 7.3% of outpatient spending.ConclusionsOur study shows the potential of health insurance claims data in identifying diseases when no diagnostic coding is available. These disease-specific spending estimates may inform Swiss health policies in cost containment and priority setting.
| S-EPMC8609863 | biostudies-literature
Project description:BackgroundMedical treatment quality has been shown to be better in high volume than in low volume hospitals. However, this relationship has not yet been confirmed in abdominal cancer in Switzerland and is relevant for referral of patients and healthcare planning. Thus, the present study investigates the association between hospital volumes for surgical resections of colon, gastric, rectal, and pancreatic carcinomas and outcomes.MethodsThis retrospective analysis is based on anonymized claims data of patients with mandatory health insurance at Helsana Group, a leading health insurance in Switzerland. Outcome parameters were length of hospital stay, mortality and cost during the inpatient stay as well as at 1-year follow-up. Hospital volume information was derived from the Quality Indicators dataset provided by the Swiss Federal Office of Public Health. The impact of hospital volume on the different treatment outcomes was statistically tested using generalized estimating equations (GEE) models, taking into account the non-independence of observations from the same hospital.ResultsThe studies included 2'859 resections in patients aged 18 years and older who were hospitalized for abdominal cancer surgery between 2014 and 2018. Colon resections were the most common procedures (n = 1'690), followed by rectal resections (n = 709). For rectal, colon and pancreatic resections, an increase in the mean number of interventions per hospital and a reduction of low volume hospitals could be observed. For the relationship between hospital volume and outcomes, we did not observe a clear dose-response relationship, as no significantly better outcomes were observed in the higher-volume category than in the lower-volume category. Even though a positive "routine effect" cannot be excluded, our results suggest that even hospitals with low volumes are able to achieve comparable treatment outcomes to larger hospitals.ConclusionIn summary, this study increases transparency on the relationship between hospital volume and treatment success. It shows that simple measures such as defining a minimum number of procedures only might not lead to the intended effects if other factors such as infrastructure, the operating team or aggregation level of the available data are not taken into account.
| S-EPMC8881861 | biostudies-literature
Project description:BackgroundA high proportion of health care services are persistently utilized by a small subpopulation of patients. To improve clinical outcomes while reducing costs and utilization, population health management programs often provide targeted interventions to patients who may become persistent high users/utilizers (PHUs). Enhanced prediction and management of PHUs can improve health care system efficiencies and improve the overall quality of patient care.ObjectiveThe aim of this study was to detect key classes of diseases and medications among the study population and to assess the predictive value of these classes in identifying PHUs.MethodsThis study was a retrospective analysis of insurance claims data of patients from the Johns Hopkins Health Care system. We defined a PHU as a patient incurring health care costs in the top 20% of all patients' costs for 4 consecutive 6-month periods. We used 2013 claims data to predict PHU status in 2014-2015. We applied latent class analysis (LCA), an unsupervised clustering approach, to identify patient subgroups with similar diagnostic and medication patterns to differentiate variations in health care utilization across PHUs. Logistic regression models were then built to predict PHUs in the full population and in select subpopulations. Predictors included LCA membership probabilities, demographic covariates, and health utilization covariates. Predictive powers of the regression models were assessed and compared using standard metrics.ResultsWe identified 164,221 patients with continuous enrollment between 2013 and 2015. The mean study population age was 19.7 years, 55.9% were women, 3.3% had ≥1 hospitalization, and 19.1% had 10+ outpatient visits in 2013. A total of 8359 (5.09%) patients were identified as PHUs in both 2014 and 2015. The LCA performed optimally when assigning patients to four probability disease/medication classes. Given the feedback provided by clinical experts, we further divided the population into four diagnostic groups for sensitivity analysis: acute upper respiratory infection (URI) (n=53,232; 4.6% PHUs), mental health (n=34,456; 12.8% PHUs), otitis media (n=24,992; 4.5% PHUs), and musculoskeletal (n=24,799; 15.5% PHUs). For the regression models predicting PHUs in the full population, the F1-score classification metric was lower using a parsimonious model that included LCA categories (F1=38.62%) compared to that of a complex risk stratification model with a full set of predictors (F1=48.20%). However, the LCA-enabled simple models were comparable to the complex model when predicting PHUs in the mental health and musculoskeletal subpopulations (F1-scores of 48.69% and 48.15%, respectively). F1-scores were lower than that of the complex model when the LCA-enabled models were limited to the otitis media and acute URI subpopulations (45.77% and 43.05%, respectively).ConclusionsOur study illustrates the value of LCA in identifying subgroups of patients with similar patterns of diagnoses and medications. Our results show that LCA-derived classes can simplify predictive models of PHUs without compromising predictive accuracy. Future studies should investigate the value of LCA-derived classes for predicting PHUs in other health care settings.
| S-EPMC8663459 | biostudies-literature
Project description:BackgroundHealth insurance claims data are used in various research fields; however, an overview on how they are used in healthcare research is scarce in Japan. Therefore, we conducted a scoping review to systematically map the relevant studies using Japanese claims data.MethodsMEDLINE, EMBASE, and Ichushi-Web were searched up to April 2021 for studies using Japanese healthcare claims data. We abstracted the data on study characteristics and summarized target diseases and research themes by the types of claims database. Moreover, we described the results of studies that aimed to compare health insurance claims data with other data sources narratively.ResultsA total of 1,493 studies were included. Overall, the most common disease classifications were "Diseases of the circulatory system" (18.8%, n = 281), "Endocrine, nutritional, and metabolic diseases" (11.5%, n = 171; mostly diabetes), and "Neoplasms" (10.9%, n = 162), and the most common research themes were "medical treatment status" (30.0%, n = 448), "intervention effect" (29.9%, n = 447), and "clinical epidemiology, course of diseases" (27.9%, n = 417). Frequent diseases and themes varied by type of claims databases. A total of 19 studies aimed to assess the validity of the claims-based definition, and 21 aimed to compare the results of claims data with other data sources. Most studies that assessed the validity of claims data compared to medical records were hospital-based, with a small number of institutions.ConclusionsClaims data are used in various research areas and will increasingly provide important evidence for healthcare policy in Japan. It is important to use previous claims database studies and share information on methodology among researchers, including validation studies, while informing policymakers about the applicability of claims data for healthcare planning and management.
| S-EPMC10407298 | biostudies-literature