Determine the therapeutic role of radiotherapy in administrative data: a data mining approach.

BACKGROUND: Clinical data gathered for administrative purposes often lack sufficient information to separate the records of radiotherapy given for palliation from those given for cure. An absence, incompleteness, or inaccuracy of such information could hinder or bias the study of the utilization and outcome of radiotherapy. This study has three specific purposes: 1) develop a method to determine the therapeutic role of radiotherapy (TRR); 2) assess the accuracy of the method; 3) report the quality of the information on treatment "intent" recorded in the clinical data in Ontario, Canada. A general purpose is to use this study as a prototype to demonstrate and test a method to assess the quality of administrative data. METHODS: This is a population based retrospective study. A random sample was drawn from the treatment records with "intent" assigned in treating hospitals. A decision tree is grown using treatment parameters as predictors and "intent" as outcome variable to classify the treatments into curative or palliative. The tree classifier was applied to the entire dataset, and the classification results were compared with those identified by "intent". A manual audit was conducted to assess the accuracy of the classification. RESULTS: The following parameters predicted the TRR, from the strongest to the weakest: radiation dose per fraction, treated body-region, disease site, and time of treatment. When applied to the records of treatments given between 1990 and 2008 in Ontario, Canada, the classification rules correctly classified 96.1% of the records. The quality of the "intent" variable was as follows: 77.5% correctly classified, 3.7% misclassified, and 18.8% did not have an "intent" assigned. CONCLUSIONS: The classification rules derived in this study can be used to determine the TRR when such information is unavailable, incomplete, or inaccurate in administrative data. The study demonstrates that data mining approach can be used to effectively assess and improve the quality of large administrative datasets.

[Clinical and organizational guidelines in university hospitals: a model for classification and analysis of hospital records]. / Atti di indirizzo clinico ed organizzativo di un'azienda ospedaliero universitaria: proposta di un modello di classificazione ed analisi.

Clinical and organizational guidelines in university hospitals: a model for classification and analysis of hospital records.

LOINC codes for hospital information systems documents: a case study.

Hospital Information Systems (HIS) handle a large number of different types of documents. Exchange and analysis of data from different HIS is facilitated by the use of standardized codes to identify document types. HL7's Clinical Document Architecture (CDA) uses LOINC (logical observation identifiers names and Codes) codes for clinical documents. The authors assessed the coverage of LOINC codes for document types in a German HIS. The authors analyzed document types that occurred more than 10 times in approximately 1.3 million documents in a commercial HIS at a major German University Hospital. Document types were mapped manually to LOINC using the Regenstrief LOINC Mapping Assistant (RELMA). Each document type was coded by two physicians. In case of discrepancies a third expert was consulted to reach consensus. For 76 of 86 document categories a LOINC code was identified, but for 38 of these categories, the LOINC code was not specific as deemed necessary. More than 93% of our local HIS documents had local document types that could be assigned a LOINC code.

On the coding and reporting of race and ethnicity in New Hampshire for purposes of cancer reporting.

BACKGROUND: The objective was to investigate how data on race and ethnicity are collected by hospitals reporting to the New Hampshire State Cancer Registry (NHSCR). METHOD: NHSCR surveyed hospitals asking how information on race and ethnicity were collected. A review of relevant legal mandates and national guidelines was undertaken. RESULTS: Many hospitals lack policies on collection, computer systems fail to support national guidelines, and staff rely on visual inspection. CONCLUSIONS: Hospital staffs are not now culturally equipped to collect race and ethnicity in a meaningful way. The numerator in cancer incidence rates is most likely not accurate and for some smaller populations very biased. A new framework is needed that takes into account the needs of the democracy.

[The necrology meeting as an instrument for quality of care improvement; experiences at the Ikazia Hospital, 1990-2000]. / De necrologiebespreking als instrument voor kwaliteitsverbetering; ervaringen in het Ikazia Ziekenhuis, 1990-2000.

The death of a patient highlights the demand for quality of care. Publication of hospital mortality figures risks incorrect interpretation and does not lead to an improvement in care. However, an above-average hospital mortality rate can be a sign of poor quality in a particular aspect of care. In the necrology meeting, the evaluation of this quality must take place with critical self-reflection, to detect opportunities by which to improve the quality of care. Quantitative data to support the necrology meeting are necessary to improve the quality. This requires systematic registration and a valid code system. In the Department of Surgery at the Ikazia Hospital Rotterdam, the Netherlands, an ABC coding system is used which indicates whether the patient died of the disease or complications, whether the death was influenced by identifiable shortcomings in the diagnostic work-up, surgical treatment or non-surgical treatment, or whether no shortcoming could be identified, and whether autopsy was carried out or refused. A more detailed registration model is developed which may be more useful in future necrology meetings and in the evaluation of these meetings.

Accuracy of centrally recorded OPCS codes for vascular surgery in the United Kingdom.

AIM: Centrally recorded OPCS codes are based upon district returns. The aim of this study is to determine the accuracy of this system with regard to vascular surgery. METHODS: Prospectively recorded audit data for vascular and endovascular procedures were compared with those obtained from the Department of Health and Welsh Office. Five U.K. hospitals were involved in the study. Data were obtained for the twelve months, 1 April 1994-30 March 1995 (these being the most up to date figures available). RESULTS: The total number of arterial reconstructions based on audit data was 1082. Those recorded by the OPCS codes were 743. This represents a discrepancy of -31.3% (range for the five hospitals: -13.1% to -63.8%). When examining specific codes similar discrepancies were seen. For example, in one hospital 38 AAA repairs were carried out but only two were centrally recorded. However, examination of ICD9 codes (relating to hospital admissions) for that hospital showed that 38 patients with AAA were admitted. A similar wide variation was seen when examining iliac and superficial femoral artery endovascular procedures. Despite the discrepancies of audit and OPCS data, the codes for reconstructions did reflect relative workload of the different hospitals. CONCLUSION: This study shows that there is a marked underestimate of vascular workload when comparing central recorded data with that obtained from local audit. Marked variation is seen in the accuracy of data submitted from different hospitals.

[The impact of the data collection method on the medico-economic classification of admissions for myocardial infarction at the public hospitals in Lyon]. / Impact du mode de recueil des données sur le classement médico-économique des séjours pour infarctus du myocarde aux Hospices Civils de Lyon.

The aim of the PMSI (Programme de Médicalisation du Système d'Information) is to describe the activity of hospitals for budget allocation. To control the quality of this information, the authors carried out a study comparing the classification in homogenous disease groups (HDG) obtained from the PMSI with that obtained from the epidemiological data base of the PRIMA trial for patients admitted to the Civil Hospitals of Lyon for myocardial infarction between September 1st 1993 and January 31st 1995. Six hundred and fifty standardised hospital summaries were reconstituted form PRIMA data and grouped using the GENRSA 3 software. Five hundred and forty-one of these hospital stays were found in the PMSI data base and grouped. The concordance not due to chance between the two classifications was then assessed by the global kappa coefficient. It was less than the 40% threshold under which concordance not due to chance is considered to be unlikely. The discordances were essentially due to the presence of an associated diagnosis classifying the hospital stay in the HDG corresponding to complicated myocardial infarction. The presence of a classifying associated diagnosis was observed significantly more often in the PRIMA than in the PMSI data base. This results in an underestimation of the hospital activity and could have important repercussions in terms of budget allocation.

Reporting on QI efforts for internal and external customers.

BACKGROUND: In April 1993 the Methodist Hospital of St Louis Park, Minnesota, released its first internal quality report on outcomes and quality improvement (QI) initiatives. When a local television news reporter mentioned the report in a segment on health care quality, public interest led the hospital to launch an annual series of external quality reports in addition to its internal quality reports. When the eight-page external report was first released in 1994, consumer response was weak, but the report generated a strong response from the mass media, trade publications, the business community, and other health care organizations nationwide. DATA COLLECTION AND USE: Data on sentinel events and outcomes analysis of a variety of clinical and administrative functions have assisted in identifying opportunities for improvement. For example, the hospital monitors the five-year survival rate for patients with myocardial infarction. With the adoption of treatment with streptokinase, data indicated frequent hypotension. Increase of infusion from 30 to 60 minutes led to a decrease in hypotension. THE REPORTS: The external report included, in shorter and simpler form, almost all the sections in the internal report, such as QI activities (teams, training, critical paths), clinical outcome measures, community health, patient satisfaction, value, and accreditation. The indicators included in the external report were selected to minimize potential misinterpretation by public audiences. CONCLUSIONS: By increasing the visibility of QI within the hospital, the internal quality reports have helped generate further QI activity, and the external report augmented further positive publicity among the local health care press. The reports are proven effective tools for communicating the hospital's ability to sustain and improve the quality of its services over time.

Prebilling DRG training can increase hospital reimbursement.

Over the past decade, appropriate reimbursement has become increasingly dependent on accurate coding and correct DRG assignment. In this article, a study outlining the cost of inaccurate coding is presented, and a program for prebilling coder training is described. The authors show that a well-designed prebilling coding program can improve the accuracy of DRG assignment and thereby increase reimbursement by 10 percent or more.

Change in the Medicare case-mix index in the 1980s and the effect of the prospective payment system.

Persistent increases in the Medicare case-mix index over the 1980s have been ascribed to changes both in medical treatment ("real changes") and in the way medical information is recorded ("coding changes") in hospitals. These changes have been attributed, in the absence of appropriate data and analyses, to the incentives of the Medicare prospective payment system (PPS). Using data for 1980-1986 from 235 hospitals, we estimate the effect on the Medicare case-mix index of a series of variables that reflect medical treatments and coding practices. Each of these underlying real or coding variables was changing prior to PPS and would likely have continued to change even in the absence of PPS. Furthermore, PPS may have had a distinct effect on these variables. These underlying trends and the PPS effects must each be estimated. Thus, the analysis begins by developing separate estimates for each of these real and coding variables (1) in the absence of PPS (autonomous effects) and (2) as a result of PPS (induced effects). Then, changes in the case-mix index are regressed against all of these variables to determine the degree to which specific autonomous real or coding variables or induced real or coding variables actually influenced measured case mix. Results show that real and coding changes each accounted for about half of the change in the Medicare case-mix index between 1980 and 1986, with the influence of coding starting to wane by 1986. PPS-induced factors explain about 80 percent of the change in measured case mix over time, autonomous factors about 20 percent. Especially powerful determinants of case-mix change included PPS-induced substitution of surgical for medical care and PPS-induced improvements in the accuracy of coding that led to assignment of patients to higher-weighted DRGs. Also, stringent Medicare peer review organizations appeared to restrain rises in case-mix indexes for their hospitals. Outpatient substitution for inpatient treatment, which others attributed to PPS, was well underway before PPS was announced.

The use of the medical record and financial data to examine the cost of infections in the elderly.

This study demonstrates what can be accomplished when the medical record and other data sources are utilized. By using the medical record as well as financial data and input from the infectious disease department, a detailed analysis of infections in an elderly population in relation to cost and length of stay was completed. This pilot study also enabled the hospital to determine the types of studies that should be done in the future. Quality assessment and improvement studies that examine the effectiveness of infection control procedures over time, the importance of examining both community-acquired and nosocomial infections, and the differences that severity of illness may have on cost and length of stay are all areas that have been identified as needing further study. Future studies in this area and other areas will continue to utilize the medical record. However, the data obtained from the medical record should not be examined alone. It should be analyzed along with other data sources such as severity of illness data, financial data, quality assessment data, infection control data, and risk management data in order to examine epidemiological trends over time. Only when several data sources are used together to investigate a particular aspect of care will that aspect of care be thoroughly and completely examined.

Accuracy of ICD-9-CM codes in identifying reportable communicable diseases.

Billing records from the outpatient clinics of a university medical center were used to identify reportable communicable diseases. Patient charts were reviewed to check the accuracy of all cases of communicable diseases not reported to the local health department. Thirty-three percent of the cases identified as one of 20 communicable diseases, using the ICD-9-CM system, were found to be incorrectly coded. This study documents a lack of specificity (numerous false positives) when using encounter form data and ICD-9-CM codes to identify communicable diseases in an outpatient setting.