[Epidemiological analysis on the hospitalization from neurological diseases in the Province of Pavia (Lombardy Region, Northern Italy)]. / Analisi epidemiologica dei ricoveri per patologie neurologiche in Provincia di Pavia.

OBJECTIVES: to evaluate time and spatial distribution of hospitalization due to neurological diseases in the province of Pavia (Lombardy Region, Northern Italy). DESIGN: ecological study. SETTING AND PARTICIPANTS: the study was performed on aggregate data of people residing in the province of Pavia in the period 2005-2014. MAIN OUTCOME MEASURES: hospital discharge records of neurological diseases and raw and standardized hospitalization rates. RESULTS: hospitalization due to neurological diseases in the Province of Pavia showed a slight decreasing trend in time. For the year 2014, the spatial analysis of hospitalizations highlights excesses of risk in the Lomellina district, both in males and in females. CONCLUSION: spatial analysis confirms previous results on specific neurological diseases and suggests more detailed analysis on hospitalization excesses in Lomellina area.

Development and representation of health indicators with thematic maps.

Italian Local Health Care Agencies (ASLs) have the role of managing the public healthcare resources in their area of competence. To this end, the ASL of Pavia has implemented a data warehouse, which collects and integrates health data of more than 500,000 people since 2004. We have exploited such data repository to compute a variety of yearly health indicators, which have been represented on thematic maps of the area. Thanks to a Web-based application, the ASL decision-makers can monitor the area with a fine-grained spatial detail, dissecting the epidemiological, economical and pharmaceutical factors underlying citizens' health and patients' care. The implemented tool is currently up-and-running and has been evaluated with a usability questionnaire on a small number of users.

Influence of initial glucocorticoid co-medication on mortality and hospitalization in early inflammatory arthritis: an investigation by record linkage of clinical and administrative databases.

BACKGROUND: While low-dose oral glucocorticoids (GCs) are recommended in the management of early arthritis, their impact on mortality is unclear. The aim of this study is to evaluate the effect of GCs on mortality in patients with early arthritis, by linking clinical and administrative databases. METHODS: The study included patients with new-onset rheumatoid arthritis (RA) or undifferentiated arthritis (2005-2010), who received DMARDs (MTX in RA or UA with poor prognosis, hydroxychloroquine in UA) and were alive at the second year of follow-up. Low-dose GCs could be prescribed. Clinical and administrative data were linked from Administrative Health Databases (AHD) of the corresponding province, which provided us with information on drug delivery, comorbidities, hospitalization, and mortality. The effect of GCs in the first year was defined using a dichotomous variable or a 3-level categorization (not delivered, ≤7.5 mg/day, or >7.5 mg/day of prednisone) on all-cause mortality, assessed with Cox regression, either crude or adjusted for age, gender, Charlson Comorbidity Index (CCI) or single comorbidities, ACPA, HAQ, and MTX in the first year. A secondary analysis of the effect of GCs on related hospitalizations (for cardiovascular events, diabetes, serious infections, osteoporotic fractures) was also carried. RESULTS: Four hundred forty-nine patients were enrolled (mean age 58.59, RA 65.03%) of which 51 (11.36%) died during the study. The median (IQR) follow-up was equal to 103.91 (88.03-126.71) months. Treatments with GCs were formally prescribed to 198 patients (44.10%) at ≤7.5 mg/day, although by the end of the study such treatments were received by 257 patients (57.24%); 88 patients (19.6%) were treated with GCs at >7.5 mg/day. In adjusted analyses, the GC delivery (HR, 95% CI 1.35 (0.74, 2.47)) did not significantly predict mortality - both at a low (HR, 95% CI 1.41 (0.73, 2.71)) and at a high (HR, 95% CI 1.23 (0.52, 2.92)) dosage. When "all-cause hospitalization" was used as an outcome, the analysis did not show a difference between patients receiving GC and patients not receiving GC. CONCLUSION: In patients with early inflammatory arthritis, the initial GC dose was higher than that prescribed by rheumatologists; however, on background treatment with DMARDs, GC treatments did not seem to increase mortality and hospitalizations.

Mining administrative and clinical diabetes data with temporal association rules.

The Regional Healthcare Agency (ASL) of Pavia has been collecting and maintaining a central data repository which stores both administrative and clinical healthcare data about the population of Pavia area. The analysis of such integrated databases could greatly help to extract useful information for the assessment of health care delivery process. In this paper we focus our attention on the care delivery flow of Diabetes Mellitus, and we show the application of an algorithm for the extraction of Temporal Association Rules on sequences of hybrid events. This method allows to properly exploit the integration of different healthcare information sources, and can be used to evaluate the pertinence of the care delivery flow for specific pathologies, in order to reassess or refine the inappropriate practices which lead to unsatisfactory outcomes.

Trends in Prescription and Determinants of Persistence to Antihypertensive Therapy : The PAPEETE Study.

OBJECTIVES: To assess trends in prescriptions, determinants and timing of treatment discontinuation and/or changes in antihypertensive drug therapy in a cohort of hypertensive patients living in Pavia, a city in the north of Italy. METHODS: The cohort included 61 493 patients aged ≥18 years who received their first antihypertensive drug prescription (monotherapy, fixed or extemporaneous combination) during the period 2003-6. Patients were classified as 'persistent' if 12 months after the beginning of treatment they were still taking a regular therapy (same drug = 'same therapy users', added one or more drugs = 'add-on therapy users', different drug = 'switchers'). Otherwise, they were classified as 'non-persistent' (stopping therapy after the first prescription = 'occasional users'; stopping treatment early = 'stoppers'; taking medicines in an erratic fashion = 'intermittent users'). RESULTS: ACE inhibitors were the most frequently prescribed drugs (22.8%), followed by ß-adrenoceptor antagonists (ß-blockers) [14.3%], diuretics (13.9%), Ca(2+) antagonists (11.4%) and angiotensin II type 1 receptor antagonists (angiotensin receptor blockers [ARBs]) [9.3%]. After 12 months, persistent patients were only 11.2% (same therapy users 6.7%, switchers 3.2%, add-on therapy users 1.3%). Non-persistent patients were 88.8% (35.3% occasional users, 20.6% stoppers, 32.8% intermittent users). Patient-related predictors of persistence were older age, male sex, concomitant treatment with antidiabetic and hypolipidaemic drugs and previous hospitalizations for cardiovascular events. Highest level of persistence was seen in patients starting with ARBs (18.8%), followed by ACE inhibitors (11.4%), ß-blockers (11.0%), Ca(2+) antagonists (10.8%) and diuretics (3.0%). Among ARBs, considering separately monotherapy and fixed-combination therapy, highest level of persistence was observed in patients starting with candesartan, irbesartan, valsartan and telmisartan given in monotherapy, and with valsartan and telmisartan given in fixed-dose combination. CONCLUSIONS: Persistence to antihypertensive treatment at 12 months is only 11.2%, being the lowest with diuretics (3.0%) and the highest with ARBs (18.8%).

Under-recognized post-stroke acute kidney injury: risk factors and relevance for stroke outcome of a frequent comorbidity.

BACKGROUND: Acute kidney injury (AKI) is emerging as a predictor of poor stroke outcome, however, it is often not recognized. The aim of our study was to evaluate post-stroke AKI burden, AKI risk factors and their influence in post-stroke outcome. METHODS: From 2013 to 2016, 440 individuals with stroke diagnosis admitted in Stroke Unit, Foundation IRCCS Policlinico San Matteo (Pavia, Italy), were retrospectively enrolled. AKI cases identified by KDIGO criteria through the electronic database and hospital chart review were compared with the ones reported in discharge letters or in administrative hospital data base. Mortality data were provided by Agenzia Tutela della Salute of Pavia. RESULTS: We included 430 patients in the analysis. Median follow-up was 19.2 months. We identified 79 AKI cases (18% of the enrolled patients, 92% classified as AKI stage 1), a fivefold higher number of cases than the ones reported at discharge. 37 patients had AKI at the admission in the hospital, while 42 developed AKI during the hospitalization. Cardioembolic (p = 0.01) and hemorrhagic (p = 0.01) stroke types were associated with higher AKI risk. Admission National Institutes of Health Stroke Scale (NIHSS, p < 0.05) and Charlson Comorbidity Index (p < 0.01) were independently associated with overall AKI, while admission NIHSS (p < 0.05) and eGFR (p < 0.005) were independently associated with AKI developed during the hospitalization. AKI was associated to longer in-hospital stay (p = 0.01), worse Rankin Neurologic Disability Score at discharge (p < 0.0001) and discharge disposition other than home (p = 0.03). AKI was also independently associated to higher in-hospital mortality (OR 3.9 95% CI 1.2-12.9 p = 0.023) but not with long-term survival. CONCLUSIONS: Post-stroke AKI diagnosis needs to be improved by strictly monitoring individuals with cardioembolic or hemorrhagic stroke, reduced kidney function, higher Charlson Comorbidity Index and worse NIHSS at presentation.

Hierarchical Bayesian Logistic Regression to forecast metabolic control in type 2 DM patients.

In this work we present our efforts in building a model able to forecast patients' changes in clinical conditions when repeated measurements are available. In this case the available risk calculators are typically not applicable. We propose a Hierarchical Bayesian Logistic Regression model, which allows taking into account individual and population variability in model parameters estimate. The model is used to predict metabolic control and its variation in type 2 diabetes mellitus. In particular we have analyzed a population of more than 1000 Italian type 2 diabetic patients, collected within the European project Mosaic. The results obtained in terms of Matthews Correlation Coefficient are significantly better than the ones gathered with standard logistic regression model, based on data pooling.

On The Correlation Between Geo-Referenced Clinical Data And Remotely Sensed Air Pollution Maps.

This work presents an analysis framework enabling the integration of a clinical-administrative dataset of Type 2 Diabetes (T2D) patients with environmental information derived from air quality maps acquired from remote sensing data. The research has been performed within the EU project MOSAIC, which gathers T2D patients' data coming from Fondazione S. Maugeri (FSM) hospital and the Pavia local health care agency (ASL). The proposed analysis is aimed to highlight the complexity of the domain, showing the different perspectives that can be adopted when applying a data-driven approach to large variety of temporal, geo-localized data. We investigated a set of 899 patients, located in the Pavia area, and detected several patterns depicting how clinical facts and air pollution variations may be related.

Integration of Administrative, Clinical, and Environmental Data to Support the Management of Type 2 Diabetes Mellitus: From Satellites to Clinical Care.

A very interesting perspective of "big data" in diabetes management stands in the integration of environmental information with data gathered for clinical and administrative purposes, to increase the capability of understanding spatial and temporal patterns of diseases. Within the MOSAIC project, funded by the European Union with the goal to design new diabetes analytics, we have jointly analyzed a clinical-administrative dataset of nearly 1.000 type 2 diabetes patients with environmental information derived from air quality maps acquired from remote sensing (satellite) data. Within this context we have adopted a general analysis framework able to deal with a large variety of temporal, geo-localized data. Thanks to the exploitation of time series analysis and satellite images processing, we studied whether glycemic control showed seasonal variations and if they have a spatiotemporal correlation with air pollution maps. We observed a link between the seasonal trends of glycated hemoglobin and air pollution in some of the considered geographic areas. Such findings will need future investigations for further confirmation. This work shows that it is possible to successfully deal with big data by implementing new analytics and how their exploration may provide new scenarios to better understand clinical phenomena.

A validation study of a new classification algorithm to identify rheumatoid arthritis using administrative health databases: case-control and cohort diagnostic accuracy studies. Results from the RECord linkage On Rheumatic Diseases study of the Italian Society for Rheumatology.

OBJECTIVES: To develop and validate a new algorithm to identify patients with rheumatoid arthritis (RA) and estimate disease prevalence using administrative health databases (AHDs) of the Italian Lombardy region. DESIGN: Case-control and cohort diagnostic accuracy study. METHODS: In a randomly selected sample of 827 patients drawn from a tertiary rheumatology centre (training set), clinically validated diagnoses were linked to administrative data including diagnostic codes and drug prescriptions. An algorithm in steps of decreasing specificity was developed and its accuracy assessed calculating sensitivity/specificity, positive predictive value (PPV)/negative predictive value, with corresponding CIs. The algorithm was applied to two validating sets: 106 patients from a secondary rheumatology centre and 6087 participants from the primary care. Alternative algorithms were developed to increase PPV at population level. Crude and adjusted prevalence estimates taking into account algorithm misclassification rates were obtained for the Lombardy region. RESULTS: The algorithms included: RA certification by a rheumatologist, certification for other autoimmune diseases by specialists, RA code in the hospital discharge form, prescription of disease-modifying antirheumatic drugs and oral glucocorticoids. In the training set, a four-step algorithm identified clinically diagnosed RA cases with a sensitivity of 96.3 (95% CI 93.6 to 98.2) and a specificity of 90.3 (87.4 to 92.7). Both external validations showed highly consistent results. More specific algorithms achieved >80% PPV at the population level. The crude RA prevalence in Lombardy was 0.52%, and estimates adjusted for misclassification ranged from 0.31% (95% CI 0.14% to 0.42%) to 0.37% (0.25% to 0.47%). CONCLUSIONS: AHDs are valuable tools for the identification of RA cases at the population level, and allow estimation of disease prevalence and to select retrospective cohorts.

Temporal data mining for the assessment of the costs related to diabetes mellitus pharmacological treatment.

Diabetes care and chronic disease management represent data-intensive contexts which allow Local Healthcare Agencies (ASL) to collect a huge amount of information. Time is often an essential component of such information, given the strong importance of the temporal evolution of the considered disease and of its treatment. In this paper we show the application of a temporal data mining technique to extract temporal association rules over an integrated repository including both administrative and clinical data related to a sample of diabetic patients. We will show how the method can be used to highlight cases and conditions which lead to the highest pharmaceutical costs. Considering the perspective of a Regional Healthcare Agency, this method could be properly exploited to assess the overall standards and quality of care, while lowering costs.