Variation in hospital performance measures from the Turkey Ministry of Health.

BACKGROUND: The Turkish healthcare system has seen broad population-based improvements in expanded health insurance coverage and access to healthcare services. Hospital performance in this national system is understudied. We aimed to identify trends in hospital performance over time following implementation of the Health Transformation Program and describe how regional outcomes correlate with regional vital statistics. OBJECTIVE: We examine hospital performance data collected by the PHA from 2013 to 2015. We aim to identify the temporal variation in hospital performance for nearly 30 individual measures and to describe the relationship between hospital-level performance measures and regional vital statistics. METHODS: We conducted a retrospective cohort study of 674 public hospitals in Turkey using baseline data from 2013 and follow-up data from 2014-15 collected by the Turkish Statistical Institution and the Public Hospital Agency. We report demographic and socioeconomic data across 12 geographic regions and analyze 29 hospital-level performance measures across four domains: (i) health services; (ii) administrative services; (iii) financial services and (iv) quality measures. We examine temporal variation, and study correlation between performance measures and regional vital statistics. We fit mixed-effects linear regression models to estimate linear trend over time accounting for within-hospital residual correlation. We prepared our manuscript in accordance with guidelines set by the STROBE statement for cohort studies. RESULTS: During the 3 years of study period, 21 of 29 measures improved and 8 measures worsened. All but three measures demonstrated significant differences across regions of the country. Several measures, including inpatient efficiency, patient satisfaction and audit score, are associated with regional infant mortality and life expectancy. CONCLUSIONS: Evidence of temporal improvement in hospital-level performance may suggest some positive changes within the Turkish national healthcare system. Correlation of some measures with regional level health outcomes suggests a quality measurement strategy to monitor performance changes in the future. Although hospital-level functions have improved performance, the results of our study may help achieve further improvement for the health of the country's citizens.

Functionality of hospital information systems: results from a survey of quality directors at Turkish hospitals.

BACKGROUND: We aimed to determine availability of core Hospital Information Systems (HIS) functions implemented in Turkish hospitals and the perceived importance of these functions on quality and patient safety. METHODS: We surveyed quality directors (QDs) at civilian hospitals in the nation of Turkey. Data were collected via web survey using an instrument with 50 items describing core functionality of HIS. We calculated mean availability of each function, mean and median values of perceived impact on quality, and we investigated the relationship between availability and perceived importance. RESULTS: We received responses from 31% of eligible institutions, representing all major geographic regions of Turkey. Mean availability of 50 HIS functions was 65.6%, ranging from 19.6% to 97.4%. Mean importance score was 7.87 (on a 9-point scale) ranging from 7.13 to 8.41. Functions related to result management (89.3%) and decision support systems (52.2%) had the highest and lowest reported availability respectively. Availability and perceived importance were moderately correlated (r = 0.52). CONCLUSION: QDs report high importance of the HIS functions surveyed as they relate to quality and patient safety. Availability and perceived importance of HIS functions are generally correlated, with some interesting exceptions. These findings may inform future investments and guide policy changes within the Turkish healthcare system. Financial incentives, regulations around certified HIS, revisions to accreditation manuals, and training interventions are all policies which will help integrate HIS functions to support quality and patient safety in Turkish hospitals.

Electronic surveillance of patient safety events using natural language processing.

OBJECTIVE: We describe our approach to surveillance of reportable safety events captured in hospital data including free-text clinical notes. We hypothesize that a) some patient safety events are documented only in the clinical notes and not in any other accessible source; and b) large-scale abstraction of event data from clinical notes is feasible. MATERIALS AND METHODS: We use regular expressions to generate a training data set for a machine learning model and apply this model to the full set of clinical notes and conduct further review to identify safety events of interest. We demonstrate this approach on peripheral intravenous (PIV) infiltrations and extravasations (PIVIEs). RESULTS: During Phase 1, we collected 21,362 clinical notes, of which 2342 were reviewed. We identified 125 PIV events, of which 44 cases (35%) were not captured by other patient safety systems. During Phase 2, we collected 60,735 clinical notes and identified 440 infiltrate events. Our classifier demonstrated accuracy above 90%. CONCLUSION: Our method to identify safety events from the free text of clinical documentation offers a feasible and scalable approach to enhance existing patient safety systems. Expert reviewers, using a machine learning model, can conduct routine surveillance of patient safety events.