What is an MPI? According to AHIMA, MPI is an acronym for master patient index. It is a list of all the known patients attending a single organization or facility whose visits are subsequently linked by a single specific identifier. Most organizations use a medical record number so that computer systems have an easier time tracking and coordinating the desired information. (AHIMA) This master patient index might list patients, health care members, subscribers, physicians, persons, payers, and any number of others.
a. Interest in electronic MPIs has increased since their initial use in the 1980. This is mostly because of the push to begin using EHR’s as well as the consolidation of many health care information exchanges. The development of …show more content…
Algorithms are discussed in the file “Fundamentals for Building a Master Patient Index/Enterprise Master Index (Updated)” provided by the instructor. Algorithms are accurate, sophisticated, flexible, mathematical formulas that are designed to be used by the organization or enterprise to combine the weighted patient data elements to aid in determining the probability of a duplication and eliminate errors. The 3 algorithms discussed are deterministic, rules based, and probabilistic. Deterministic- deemed to be about 20-40% accurate with patient identification and they usually have the result of many false matches. Very poor at matching people when numbers or letters are transposed, names are changed, there is a large amount of data, or limited data. Rules based algorithms- use a more sophisticated method. These type systems are often called “fuzzy logic”. This method lets the organization assign weights to specific properties they are interested in and then use these weights to compare specific records to each other as desired. In order for this system to properly function, the organization has to estimate the weights ahead of time then use those weights in the analysis. Accuracy using this system can vary greatly and only 50-80% of the duplicate records may be identified requiring multiple repetitions. Probabilistic- This system has an accuracy rate of over 90% and is considered to be the most advanced available. They are based on formula that can analyze site-specific …show more content…
Probabilistic is considered to be the best algorithm because it can be set to analyze the data at the specific site to weigh the desired information and it is more sophisticated to allow for the handling of much larger sets of data. This particular system has an accuracy rate of over 90% and is considered to be the most advanced available. It has advanced search capabilities that allow for alternate name cross-referencing, digit transposition and rotations, and much
a. Interest in electronic MPIs has increased since their initial use in the 1980. This is mostly because of the push to begin using EHR’s as well as the consolidation of many health care information exchanges. The development of …show more content…
Algorithms are discussed in the file “Fundamentals for Building a Master Patient Index/Enterprise Master Index (Updated)” provided by the instructor. Algorithms are accurate, sophisticated, flexible, mathematical formulas that are designed to be used by the organization or enterprise to combine the weighted patient data elements to aid in determining the probability of a duplication and eliminate errors. The 3 algorithms discussed are deterministic, rules based, and probabilistic. Deterministic- deemed to be about 20-40% accurate with patient identification and they usually have the result of many false matches. Very poor at matching people when numbers or letters are transposed, names are changed, there is a large amount of data, or limited data. Rules based algorithms- use a more sophisticated method. These type systems are often called “fuzzy logic”. This method lets the organization assign weights to specific properties they are interested in and then use these weights to compare specific records to each other as desired. In order for this system to properly function, the organization has to estimate the weights ahead of time then use those weights in the analysis. Accuracy using this system can vary greatly and only 50-80% of the duplicate records may be identified requiring multiple repetitions. Probabilistic- This system has an accuracy rate of over 90% and is considered to be the most advanced available. They are based on formula that can analyze site-specific …show more content…
Probabilistic is considered to be the best algorithm because it can be set to analyze the data at the specific site to weigh the desired information and it is more sophisticated to allow for the handling of much larger sets of data. This particular system has an accuracy rate of over 90% and is considered to be the most advanced available. It has advanced search capabilities that allow for alternate name cross-referencing, digit transposition and rotations, and much