Linkage Methods Research Articles

Linkage of national clinical datasets without patient identifiers using probabilistic methods.

ObjectivesTo develop a step-by-step process for probabilistic linkage of national clinical and administrative datasets without personal information, providing guidance on selecting variables for linkage, estimating match weights, and choosing the probabilistic linkage threshold. To validate this process against deterministic linkage using patient identifiers.
 ApproachWe undertook probabilistic linkage without personal information using electronic health records from the National Bowel Cancer Audit (NBOCA) and Hospital Episode Statistics (HES) databases for bowel cancer patients undergoing emergency surgery in England. We selected linkage variables based on completeness, and ability to discriminate between matches and non-matches, assessed using a novel score derived from m-probabilities and u-probabilities. Taking deterministic linkage using patient identifiers as the reference-standard, we calculated sensitivity and specificity of probabilistic linkage, plotted a Receiver Operating Characteristic curve across alternative thresholds of match weights, and compared patient characteristics and estimates from fitted regression models between linkage methods.
 ResultsWhen considering the ability to discriminate between matches and non-matches, patient and administrative variables tended to discriminate better than clinical variables. 81.4% of NBOCA records were linked to HES using probabilistic linkage, versus 82.8% using deterministic linkage. Most NBOCA records were linked to HES using both methods (8,427/10,566). Probabilistic linkage had over 96% sensitivity and 90% specificity compared to deterministic linkage using patient identifiers. Patients that linked deterministically, but not probabilistically, were younger and more likely to have emergency admission, but otherwise had similar characteristics. Regression models for mortality and length of hospital stay according to patient and tumour characteristics were not sensitive to the linkage approach.
 ConclusionProbabilistic linkage without personal information can be used as an alternative to deterministic linkage using patient identifiers, or as a method for enhancing deterministic linkage. It allows analysts outside highly secure data environments to undertake linkage while minimising costs and delays, protecting data security, and maintaining linkage quality.

Linking Administrative Test and Trace Data to the Covid-19 Infection Survey to Improve Analysis of Positivity and Vaccine Effectiveness.

The COVID-19 Infection Survey (CIS) allows us to understand many aspects of the pandemic. To enable more in-depth accurate analysis on positivity and vaccine effectiveness we have needed to link the survey at pace to NHS Test and Trace (T&T) data to better inform policy decisions at the highest level. The complex nature of the T&T dataset has been challenging to the development of a robust linkage method capable of producing regular timely updates. The data first required extensive cleaning, editing and standardisation. A combination of deterministic, probabilistic, and associative linkage techniques was used to produce four pairwise linkages: T&T was linked to itself to create person-level test clusters and a unique person ID; CIS-Personal Demographic Service (PDS) and T&T-PDS linkage tables were then produced assigning NHS number to maximise the quality of the linkage between the datasets, before producing a final separate anonymised CIS-T&T linked product fit for analysts. Post-linkage quality assessment to estimate the number of false positives (false links) and false negatives (missed matches) was undertaken and utilised in the context of true positives to estimate precision (accuracy) and recall (coverage) of the linked method, both found to be over 98% for the CIS-T&T linkage. Continuous methodological development has significantly improved the linkage quality and match rate over time. The additional test data provided by the linkage enables analysts to identify participants testing positive between their scheduled monthly visits. This adds key missing information, enabling the level of protection from re-infection afforded by previous infection to be estimated with increased accuracy and reduced bias. Analysts can now access enriched linked datasets with confidence that the linkage method meets the required quality standards. The analysis produced from this linked data has provided crucial evidence to government on re-infection and infection following vaccination, which has enabled more targeted public health policies to help avoid unnecessary economic harm. Widely published statistics have helped inform the public’s understanding of the pandemic and their personal risk.

COVID-19 transmission and infection: linkage of COVID-19 Infection Survey, Test and Trace, and Patient Demographics Survey.

ObjectivesData linkage was conducted between the Office for National Statistics’ Covid Infection Survey (CIS), the Department of Health and Social Care’s Test and Trace (T&T) and NHS’ Personal Demographics Service (PDS) datasets. Linked data was required to provide reliable estimates of rates of COVID-19 transmission and infection used to inform policy regarding the ongoing pandemic.
 ApproachThe CIS was created to track infection rates in the UK population. Linking CIS participants to positive tests in T&T helped improve these estimates. Linkage to PDS was required to attach NHS number to these datasets to facilitate further linkages that could also be used to inform Government about the spread of the virus. Multiple approaches were used to link the data. Initially, T&T was linked to itself via a series of strict matchkeys to cluster records belonging to the same individual, to create a person level identifier. Subsequent linkage of CIS-PDS, T&T-PDS and CIS-T&T involved deterministic linkages with matchkeys designed and applied independently. A probabilistic (Fellegi-Sunter scoring) method was used to link CIS-PDS and CIS-T&T. Additional, associative links were created between CIS and T&T records that had matched to the same PDS record but had not matched to each other.
 ResultsThe accuracy of CIS-PDS and CIS-T&T linkages was high (recall and precision >98%; all 95% lower confidence intervals >93%). A quality assessment of T&T-PDS is underway, as are relevant bias analyses.
 ConclusionAs a result of this linkage, COVID-19 analysts have access to enriched datasets linked to compare previously separated variables, with confidence that the linkage method used was to required quality standards. The linked data has been used to provide crucial evidence to Government on infection and re-infection rates. Subsequent linkages have enabled analysts to explore risk factors associated with different variants of the virus, vaccination status and hospital episodes. Improvements continue to be made.

Bespoke automated linkage to enable analysis of covid deaths by ethnicity.

In early 2020 there was intense media speculation that ethnicity and Covid-19 deaths were correlated. However, the existing method of adding ethnicity to death records resulted in low linkage rates for very recent deaths. We designed and implemented a bespoke linkage in three days enabling accurate reporting to the nation.
 We linked the 2011 England and Wales Census to death records using a range of personal identifiers. Due to time pressure, we focused on executing a single linkage method well. Deterministic linkage was chosen, using a variety of matchkeys which were tested via clerical review. To overcome the issue of addresses changing since 2011, we also linked 2020 death record residuals to the 2019 Patient Register (PR) and then made use of the 2011 PR address where it existed. This additionally provided an indication of whether unmatched death records might be attributable to migration into England and Wales post-2011.
 The prior linking method used NHS Number only. Although the overall linkage rate was approximately 90%, the rate for recent deaths (2nd March 2020 to 10th April 2020 in the first iteration of the linkage) was closer to 30% due to an administrative lag in adding NHS Numbers to death records. Our novel bespoke linkage method linked over 39,000 extra death records. Whilst this had minimal impact on the overall linkage rate, it improved the linkage rate for recent deaths to approximately 90%. This was without an impact on accuracy: clerical review demonstrated that the false positive rate was approximately 0.2%. A report was published using this data showing that the risk of death involving Covid-19 among some ethnic groups was significantly higher than others.
 Determining whether Covid-19 disproportionally affected certain ethnicities was of crucial importance in the early phase of the pandemic to enable appropriate government strategies to be developed. We delivered a bespoke linkage under an exceptional time-limit without compromising on accuracy, enabling this impactful analysis with nation-wide interest and impact.

Permanent Magnet Flux Linkage Identification Method of the Model Predictive Current Control System for PMSM

Accurate identification of permanent magnet flux linkage can effectively improve the control performance of model predictive current control system of permanent magnet synchronous motor (PMSM). This paper reviews the existing identification methods of permanent magnet flux linkage of PMSM first and then proposes an identification method of permanent magnet flux linkage based on an improved unscented particle filter, the problem of particle diversity reduction of the traditional unscented particle filtering algorithm is solved, and the identification performance of permanent magnet flux is improved. Simulation research and experimental verification of the proposed method are carried out, and the simulation and experimental results confirm the effectiveness of the proposed method.

Access to Cervical Cancer Screening for Women Living With HIV: A Record Linkage Study.

BackgroundAccording to the Global Strategy to Accelerate the Elimination of Cervical Cancer launched by the World Health Organizations (WHO), 70% of all women should undergo cervical cancer (CC) screening by 2030. While integrated HIV care and CC screening services exist, lack of unified data and patient management systems hinder reaching the set WHO goals. We used record linkage methods to assess the number of women receiving HIV care and visual inspection with acetic acid (VIA) based CC screening at Kanyama First Level Hospital (FLH) in Lusaka, Zambia.
 MethodsWe retrieved HIV patient records from SmartCare, Zambia’s National Electronic Health Records system, and the CC screening database for the years 2004 to 2018, and 2010 to 2018, respectively. We restricted this pilot study to women attending Kanyama FLH for HIV services, and its surrounding clinics for CC screening. We used the following linkage variables: first name, last name, facility, and year of birth. We used a machine learning based approach with Random Forest classifiers to identify records from the HIV and CC screening services belonging to the same person. We created synthetic patient record datasets to train and test the linkages. After the linkage, we anonymized the data and retrieved screening status from the CC screening database. We assessed the percentage of women in HIV care undergoing CC screening and the respective screening results. We calculated the time to CC screening with reference to the start of the HIV Antiretroviral Therapy treatment (ART).
 ResultsWe retrieved a total of 32,970 records from SmartCare and 176,736 records from CC screening databases. With the linkage we determined that of all women that ever accessed HIV treatment services at Kanyama FLH between 2004 and 2018, 2,999 (9%) received CC screening. Of those that were screened, 30% received the CC service at Kanyama FLH, while the rest from 13 surrounding facilities. 61% of women accessed CC services before starting ART. Of the women screened for CC, 371 had a positive VIA result (12%), 2,395 (80%) a negative and 233 (8%) had a missing or indeterminate result. When we restricted the analysis to women, who were eligible for screening (aged 25 to 59 years), the screening coverage was 9.6%.
 ConclusionOur study demonstrates the significance of record linkage efforts in determining CC screening coverage among women living with HIV. According to these findings, CC screening coverage among patients receiving ART services between 2004 and 2018 was at 9%, showing the depth of effort required to meet WHO targets at the time. The linkage output was primarily influenced by the scarcity of available linkage variables and data missingness, resulting in potentially overestimating or underestimating the CC coverage. This is typical of data systems in low- and middle-income countries; however, this study demonstrates that record linkage is useful in bridging these widespread data gaps.

Identifying prenuptial births from family pedigrees using record linkage.

ObjectivesDemographers are interested in the degree to which marriage is driven by prenuptial pregnancy within particular communities. To help answer this question, we present a novel method which links marriage certificates to birth certificates, where the birth-mother is the marriage-bride, considering only births which occur in the first seven months after a marriage. ApproachTo identify prenuptial births we employed an unsupervised graph-based record linkage method to link birth and marriage certificates. We first extracted related groups of individuals: babies and their parents from birth certificates, and brides, grooms, and their parents from marriage certificates. To link births with marriages, we employed techniques to address challenges such as changing attribute values over time (such as names and addresses), the ambiguity of attribute values giving priority to rare names over common names, and different relationships encountered at different points in time (by applying temporal constraints). Based on the obtained links we then extracted prenuptial births. ResultsUsing two Scottish data sets containing a total of 38,451 births and 8,667 marriage certificates from the period 1861 to 1901 and employing different linkage thresholds we identified between 853 and 945 first birth-marriage links in the smaller (rural) data set, and between 2,165 and 2,232 links in the larger (urban) data set. In the rural data set, between 16.9% and 17.7% of these links were with birth less than 8 months after marriage (i.e. prenuptial births), where the corresponding ground truth contained 17.6% prenuptial births. For the urban data set, we identified between 51.3% and 51.4% prenuptial births. Our results show clear differences between rural and urban prenuptial pregnancies in 19th Century Scotland. ConclusionWe have presented an unsupervised graph-based record linkage method that compares attribute values of individuals and their relationships to link records based on the ambiguity of attribute values, attribute value changes, and temporal constraints. This linking process helps us to identify prenuptial births to help us understand the degree to which marriages may have been driven by pregnancy.

Making the 2022 Rwanda Census count.

For the 2022 Rwanda Census, data will be collected electronically for the first time. This provides both an opportunity and a challenge to use technology for census processing. We are working with National Institute of Statistics Rwanda (NISR) to help overcome linkage problems and produce robust population estimates. Working with NISR we aim to create an achievable Census solution given the constraints on time, computing power and experience. We identified key issues, such as the lack of addresses and full date of birth, and developed a series of recommendations. These combined our experience of census processing and NISR’s knowledge of Rwandan geography, conventions and culture to develop the simplest solutions possible, ensuring that each recommendation gave maximum benefit for the effort required. Next, we plan to use 2021 Rwanda Census Pilot data to begin development of an end-to-end matching exercise including pre-processing, automatic linkage, clerical resolution and quality assurance. We developed recommendations, including small changes to the data collection application and creating a Unique Property Reference Number by concatenating the Enumeration Area code and the Structure Number written on the building during Census collection. Following this, we produced a synthetic Rwandan Census and Post Enumeration Survey (PES) which was used to demonstrate and collaboratively develop matching algorithms. Probabilistic linkage methods were considered but discarded due to lack of sufficient infrastructure. Instead, deterministic methods were used. Match keys are currently being developed by NISR with clerical review undertaken using our open-source Clerical Matching Online Widget (CROW) with the synthetic data. Future plans include using the Census Pilot to estimate overcount and developing a modified version of the Soundex phonetic encoder for use on Rwandan names. The resource available for the Rwandan Census is comparatively small, but even the smallest resource can have a massive impact when used productively. This collaborative project takes the key concepts of well-developed linkage strategy and makes appropriate modifications to enable an achievable and effective Rwandan Census linkage strategy.

A proposed approach for standardized reporting of data linkage processes and results.

ObjectivesPopulation Data BC (PopData) is an agency in British Columbia, Canada, that routinely performs linkages of various administrative and researcher-collected data to a population spine. We developed a linkage report template in order to increase transparency of linkage process and outcome for end users and data providers. ApproachPopData performs probabilistic and deterministic data linkage using an in-house software. A literature review identified existing guidelines and examples of linkage reporting. A survey collected input from a wide range of end users about their interest in receiving linkage reports and specific information that is important to their work. A draft template was developed by PopData’s linkage experts and data scientists which then was reviewed by PopData staff and external partners. Privacy requirements, mode of delivery, readability to the intended audience and operational feasibility were carefully considered. ResultsThe resulting template built on our existing internal linkage summaries. The report follows a framework suggested in the literature with three key components: 1) information on the data source and linkage fields, 2) data pre-processing and linkage methodology, and 3) linkage results, presented in tables and figures, including overall linkage rates, detail on matched fields, and the distribution of linkage weights of linked and unliked pairs. In addition, an appendix describes the linkage methods and population spine in detail, and supplementary notes will comment on unique issues related to the data, when those are applicable. Educational materials to aid understanding of linkage methodologies and reporting are also under development. ConclusionLinked data are increasingly used in research, making it important to provide information on linkage process and performance to the research community. Rigorous and standardized linkage reports produced by data centres can facilitate evaluation of the impact of linkage performance on research findings and enable transparent reporting in peer-reviewed research.

Analysing linked 2021 Census and Admin data to inform population transformation statistics.

ObjectivesTo analyse the linked dataset between England and Wales 2021 Census/Census Coverage Survey and the Demographic Index (DI), would provide insights on quality and coverage, enabling improvements of using admin data to form population/ household estimates for future linkages and to evidence additional admin sources needed to capture the population. ApproachThe DI is a composite dataset made up of several linked admin sources. Therefore, the linked output is complex, made up of clusters where individual sources within the clusters have been linked to the Census using the different available information across sources. To carry out the analysis, a flagging strategy has been designed to enable analysts to form cuts of the linked dataset that are specific to their research needs. High level research questions have been designed to provide fast paced analysis to inform the National Statistician’s 2023 recommendation on the need for a future Census. ResultsThe results will inform the National Statistician’s 2023 recommendation, but also the future delivery of transformed population and migrations statistics . This includes: Providing insight into how to improve the current linkage methodology of the DI How we use the DI to construct Statistical Population Datasets (which use inclusion rules to establish the usual resident population) Identifying the extent of over and under coverage in the Statistical Population Datasets, allowing development of an appropriate estimation strategy to more accurately estimate the population The results will also inform how we utilise linked composite data sources in the future, as we can share the lessons learnt in how we planned to use the data, including our flagging strategy verses the reality faced when undertaking the analysis. ConclusionThis analysis will inform approaches to using linked composite data in the future but also provide a wealth of knowledge including: informing linkage methods, defining the population estimation challenge and providing insights into how we deliver admin based population estimates.

Ancillary Data Record Linkage to characterize the completeness of data for the All of Us Research Program.

ObjectiveThe All of Us Research Program (AoURP) is an ambitious effort to gather health data from one million Americans to accelerate research. We linked Electronic Health Records (EHR) and insurance claims data to characterize the degree to which ancillary datasets can improve data completeness for care received by AoURP participants. ApproachWe sought to link EHR data for 400,000 consented AoURP participants with insurance claims data provided by IPM.AI (Swoop Analytics), a commercial analytics company who have insurance claims data for 300M (over 90% of) Americans. We utilized a HIPAA-compliant privacy-preserving record linkage method (tokenization, provided by Datavant) to match patients between datasets. We evaluated match fidelity and the degree of overlap between AoURP EHRs and IPM.AI claims data. We characterized the association of patient and organizational level factors (demographics, healthcare provider organization, reporting site) with match performance. ResultsAs of submission of this abstract, 41% of AoURP EHRs matched with IPM.AI claims. We compared patient healthcare encounters, diagnosis codes (DX), procedure codes (PX), and national drug codes (NDC) for matched patients by month. The union of AoU and IPM.AI data greatly increased data completeness in matched patients. Only 20% of healthcare encounters were seen by AoURP and IPM.AI concurrently while 25% were unique to AoU EHRs and 55% to IPM.AI claims on a monthly level. The number of diagnosis events compared between AoURP and IPM.AI is roughly equal (AoU +6%) while procedure events are elevated in claims data (23%) and drug counts are greatly elevated in AoURP EHR data (71%). We found that matched patients had more healthcare encounters compared to unmatched patients. ConclusionTo our knowledge this is the first effort to address challenges in AoURP data completeness through complementary data linkage. Our results suggest that supplementary data linkage can improve data completeness in a large national research initiative. We identified several patient factors that require further investigation in improving match fidelity.

Comparison of the Use of Linkage in Cluster Integration With Path Analysis Approach

This study aims to compare integrated clusters on various linkages (Ward, Complete, and Average linkage) with path analysis to classify the behavior of subservient to pay for mortgages. The data used are primary data from questionnaires filled out by Bank X customers, using purposive sampling. The measurement method used is the average score method. Liu et al. with the title “Enhanced path-based Clustering Algorithm” shows the results that path analysis with cluster integration gives results that outperform other cluster algorithms. So, this study uses the cluster integration with path analysis method by comparing the linkage. The results showed that cluster integration in the Ward linkage method with path analysis was the best method, by comparing the coefficient of determination. The variables of attitude, subjective norms, behavioral control, and intention to comply with paying explained the behavior of Bank X customers to comply with paying mortgages by 88.32%, while the remaining 11.68% was explained by variables outside the model.

Effects of Soil Map Scales on Estimating Soil Organic Carbon Stocks in Southeastern China

Digital soil maps of different scales have been widely used in the estimates of soil organic carbon (SOC). However, exactly how the scale of the soil map impacts SOC dynamics and the key factors influencing SOC estimations during the map generalization process have rarely been assessed. In this research, a newly available soil database of Zhejiang Province in southeastern China, which contains 2154 geo-referenced soil profiles and six digital soil maps at scales of 1:50,000, 1:250,000, 1:500,000, 1:1,000,000, 1:4,000,000, and 1:10,000,000, and three different linkage methods (i.e., the mean, median, and pedological professional knowledge-based (PKB) methods) were used to evaluate their influence on the estimates of SOC. The findings of our study were as follows: (1) The scale of the soil map was identified as being of crucial importance for regional SOC estimations. (2) The linkage method played an important role in the accurate estimates of SOC, and the PKB method could provide the most detailed information on the spatial variability of SOC estimations. (3) The key factors affecting the estimates of SOC during the map generalization process as the soil map scale decreased from 1:50,000 to 1:10,000,000 were determined, including the changes in the number of soil profiles, the conversions between different soil types, the conversions from non-soils to soils, and the linkage methods of aggregating the SOC density values of soil profiles to represent map units. The results suggest that the most detailed 1:50,000-scale soil map coupled with the PKB method would be the optimal choice for regional SOC estimations in China.

Validation of using disqualification from long-term care insurance registry as a proxy of death

Objectives The national database for long-term care insurance (LTCI) of Japan (Kaigo DB) enables researchers to access comprehensive data from its LTCI registry, eligibility assessment records, claims for service usage, and information about service providers. However, studies regarding the death or mortality of beneficiaries cannot be conducted because Kaigo DB does not contain death records, and researchers are not allowed to link Kaigo DB to other databases, such as national death records. Therefore, we aimed to assess the validity of using an insurer's disqualification from an LTCI beneficiary as a proxy of death.Methods We used 510,751,798 monthly beneficiary records between April 2007 and March 2017 from the LTCI registry, while excluding data for ineligible persons for LTCI benefit or those younger than 65 years. We identified insurer cases disqualified from LTCI beneficiaries and linked them to national death records using deterministic linkage methods by dates of birth and death, sex, and residence. We considered the cases as positive if they were disqualified and their record was linked to a death. We used sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) as validity indices.Results We identified 5,986,991 (1.17%) disqualified and 5,295,961 death cases. Sensitivity, specificity, PPV, and NPV of disqualification for death were 100%, 99.86%, 88.46%, and 100%, respectively. After stratification, PPV of disqualification was between 85% and 88% before 2012, 91% after 2012, 91.9% in men, and 85.9% in women. PPV increased with age (65-69 years: 80.6%, 70-74 years: 86.7%, 75-79 years: 86.4%, 80-84 years: 86.7%, 85-89 years: 88.0%, 90-94 years: 90.6%, and 95+ years: 93.4%) and level of care needed (support level: 72.2%, care level (CL) 1: 79.7%, CL2: 85.9%, CL3: 89.3%, CL4: 92.3%, and CL5: 94.0%).Conclusions Disqualification from the LTCI registry is an inappropriate measure to estimate mortality accurately because it has a 10% false-positive rate. However, it appears sufficiently valid to use disqualification as a proxy outcome of death, although the main effect or confounding of a possible predictor of death could be slightly underestimated.

Chemometric Approach to the Elemental Composition of Ethereal Oil Plant Materials

The aim of the present study is the effective application of chemometric approach to the elemental composition of a data set of ten ethereal oil plant materials. The contents of 28 essential and toxic mineral elements at main, minor and trace level are determined in each sample by means of direct solid sampling electrothermal vapourization inductively coupled plasma optical emission spectrometry (ETV-ICP-OES) and are subjected to hierarchical and non-hierarchical cluster analysis and principal component analysis. Z-transformation of the raw data, Ward’s method of linkage and squared Euclidean distance as similarity measure are used in order to discover appropriate linkage between the plant materials and between the chemical parameters characterizing them. Thus, the investigated plant samples are classified, depending on their chemical composition. On the other hand, the mineral elements are categorized into several patterns which are related to the specific soil characteristics.

Identifying and validating subtypes of Parkinson's disease based on multimodal MRI data via hierarchical clustering analysis.

ObjectiveWe wished to explore Parkinson's disease (PD) subtypes by clustering analysis based on the multimodal magnetic resonance imaging (MRI) indices amplitude of low-frequency fluctuation (ALFF) and gray matter volume (GMV). Then, we analyzed the differences between PD subtypes.MethodsEighty-six PD patients and 44 healthy controls (HCs) were recruited. We extracted ALFF and GMV according to the Anatomical Automatic Labeling (AAL) partition using Data Processing and Analysis for Brain Imaging (DPABI) software. The Ward linkage method was used for hierarchical clustering analysis. DPABI was employed to compare differences in ALFF and GMV between groups.ResultsTwo subtypes of PD were identified. The “diffuse malignant subtype” was characterized by reduced ALFF in the visual-related cortex and extensive reduction of GMV with severe impairment in motor function and cognitive function. The “mild subtype” was characterized by increased ALFF in the frontal lobe, temporal lobe, and sensorimotor cortex, and a slight decrease in GMV with mild impairment of motor function and cognitive function.ConclusionHierarchical clustering analysis based on multimodal MRI indices could be employed to identify two PD subtypes. These two PD subtypes showed different neurodegenerative patterns upon imaging.

Mastering linked datasets: The future of emergency health care research

Objectives The aim of this work is to describe routine integration of prehospital emergency health records into a health master linkage file, delivering ongoing access to integrated patient treatment and outcome information for ambulance-attended patients in Queensland. Methods The Queensland Ambulance Service (QAS) data are integrated monthly into the Queensland Health master linkage file (MLF) using a linkage algorithm that relies on probabilistic matches in combination with deterministic rules based on patient demographic details, date, time, and facility identifiers. Each ambulance record is assigned an enduring linkage key (unique patient identifier) and further processing determines whether each record matches with a corresponding hospital emergency department, admission, or death registry record. In this study, all QAS electronic ambulance report form (eARF) records from October 2016 to December 2018 where at least one key linkage variable was present (n = 1,771,734) were integrated into the MLF. Results The majority of records (n = 1,456,502; 82.2%) were for transported patients, and 90.1% (n = 1,312,176) of these transports were to public hospital facilities. Of these transport records, 93.9% (n = 1,231,951) matched to ED records and 59.3% (n = 864,394) also linked to admitted patient records. Of ambulance non-transport records integrated into the MLF, 23.6% (n = 74,311) matched with ED records. Conclusion This study demonstrates robust linkage methods, quality assurance processes, and high linkage rates of data across the continuum of care (prehospital/emergency department/admitted patient/death) in Queensland. The resulting infrastructure provides a high-quality linked dataset that facilitates complex research and analysis to inform critical functions such as quality improvement, system evaluation, and design.

Use of grouping techniques applied to the numbers of infected by Covid-19 in the Brazilian States comparing with the HDI of each State

This article aims to show how a cluster analysis can be carried out, through the technical hierarchy and the non-hierarchy in the rate of infected by Covid-19 of the Brazilian states through the numbers of infected states in order to identify a similarity between the states. through the numbers of infected, offering a counterpoint to the criterion used to analyze the number of infected in the states, based on the size of the population and comparing with their Human Development Index (HDI). public and free platforms called Coronavirus//Brasil and Atlas Brasil 2013 in relation to the 2010 HDI. For the cluster analysis, the Mahalanobins matrix was used with the hierarchical method, the simple link, complete, average, ward link and a non-hierarchical method using the K-means method were also applied, the coefficient of confenetic correlation to measure the degree of fit between the original similar matrices and the matrix resulting from the simplification provided by the clustering method. However, the method that best represents the data was found to be the complete linkage method. When grouping the states of both data, it took into account the similarity between the variables of the data and the correlation where it can be observed that the data are correlated.

Relationship between patient experience and hospital readmission: system-level survey with deterministic data linkage method

BackgroundLinkage of public healthcare data provides powerful resources for studying from a comprehensive view of quality of care than information for a single administrative database. It is believed that positive patient experiences reflect good quality of health care and may reduce patient readmission. This study aimed to determine the relationship between patient experience and hospital readmission at a system level by linking anonymous experience survey data with de-identified longitudinal hospital administrative admissions data.MethodsData were obtained by linking two datasets with anonymised individual-level records from seven largest-scale acute public hospitals over seven geographical clusters in Hong Kong. Selected records in the two datasets involving patient experience survey (PES) (2013 survey dataset) and healthcare utilization (admissions dataset) were used. Following data cleaning and standardization, a deterministic data linkage algorithm was used to identify pairs of records uniquely matched for a list of identifiers (10 selected variables) between two datasets. If patient’s record from the survey dataset matched with the hospitalization records in the admissions dataset, they were included in the subsequent analyses. Bivariate analyses and multivariable logistic regression models were performed to evaluate the associations between hospital readmission in the next calendar month and patient experience.ResultsThe overall matching rate was 62.1% (1746/2811) for PES participants aged 45 or above from the survey dataset. The average score for overall inpatient experience was 8.10 (SD = 1.53). There was no significant difference between matched patients and unmatched patients in terms of their score for the perception of overall quality of care received during hospitalization (X2 = 6.931, p-value = 0.14) and score for overall inpatient experience (X2 = 7.853, p-value = 0.25). In the multivariable model, readmission through the outpatient department (planned admission) in the next calendar month was significantly associated with a higher score given to the overall quality of care received (adjusted OR = 1.54, 95%CI = 1.09–2.17), while such association was absent for readmission through Accident and Emergency department (adjusted OR = 0.75, 95%CI = 0.50–1.12).ConclusionsThis study demonstrated the feasibility of routine record linkage, with the limited intrusion of patients’ confidentiality, for evaluating health care quality. It also highlights the significant association between readmission through planned readmission and a higher score for overall quality of care received. A possible explanation might be the perceived better co-ordination between outpatient departments and inpatient service and the well-informed discharge plan given to this group of patients.

Cluster Analysis on Various Cluster Validity Indexes with Average Linkage Method and Euclidean Distance (Study on Compliant Paying Behavior of Bank X Customers in Indonesia 2021)

This study aims to examine the differences in various cluster validity indexes in the grouping of credit customers at Bank X Malang City, Indonesia using the average linkage and Euclidean distance methods. This study uses primary data with the variables used are service quality, environment, mode, willingness to pay, and obedient paying behavior obtained through a questionnaire with a Likert scale through purposive sampling distributed to 100 respondents. The data are then analyzed by clusters using the ward linkage and Euclidean distance methods on various validity cluster indexes, including the Silhouette Index, Krzanowski-Lai, Dunn, Gap, Davies-Bouldin, Index C, Global Sillhouette, Goodman-Kruskal in this study used as a tool analysis. This study uses R software. The results show that the Krzanowski-Lai, Dunn, Gap, Global Sillhouette, and Goodman-Kruskal indices have the same cluster members, as well as the Silhouette and Davies-Bouldin indices. The best cluster indexes are the Silhouette and Davies-Bouldin indexes. All validity indices produce variance between and within the same cluster. The novelty of this study is to compare 8 validity indices, namely Sillhouette Index, Krzanowski-Lai, Dunn, Gap, Davies-Bouldin, Index C, Global Sillhouette, and Goodman-Kruskal simultaneously.