HIM 220 Module 6 Data Governance Short Paper Example

Someone Has to Own It: Data Governance and Patient Matching at Kettle River Health

[Student Name]

Southern New Hampshire University

HIM 220: Healthcare Data Management

Module Six Short Paper

[Instructor Name]

[Date]

The organization, setting and figures below are a composite written as a model document. No real employer, client, colleague or patient is described.

Someone Has to Own It: Data Governance and Patient Matching at Kettle River Health

Kettle River Health's data problems have one thing in common: nobody owns them. Registration creates duplicate records, but the HIM department cleans them up. Quality and finance define readmissions differently because no one had authority to choose. Race and ethnicity questions are asked differently at each site. This paper proposes a governance structure and applies it to the system's most serious data problem, patient matching.

Governance Versus Management

Data governance decides who has authority over data and how decisions are made: which definitions are official, who may change them, who can access what and how disputes are settled. Data management carries out those decisions through daily work such as loading the warehouse, merging duplicate records and maintaining code mappings. Kettle River has a capable data management staff but almost no governance, which is why good work in one department is undone by practices in another.

The Governance Structure

The proposed structure has four layers. At the top, a governance committee led by the system's chief operating officer sets policy and breaks deadlocks. Data owners, senior leaders accountable for a domain such as patient demographics or clinical results, approve definitions in their domain. Data stewards, experienced staff who know how data are created and used, write definitions, monitor quality and propose fixes. Data custodians in information technology maintain the systems and apply the rules technically.

Table 1. Governance Roles at Kettle River Health

Note. Proposed by the author for committee approval.

Decision Rights and Escalation

Each data element in the dictionary will list its owner and steward. Stewards may make routine changes, such as adding an allowed value, with owner approval. Changes affecting more than one domain or any board-reported measure go to the committee. When departments disagree, stewards first try to reconcile definitions; unresolved disputes go to the committee within one meeting cycle. This structure would have resolved the readmission dispute months earlier.

Measuring Quality Under Governance

Governance needs measurements. Stewards will borrow the three check families from Kahn et al. (2016), namely conformance, completeness and plausibility, and will report a quarterly scorecard using those categories for their domains. The committee will set targets, such as reducing suspected duplicates below 2% and missing race and ethnicity below 5%, and review progress at each meeting.

Why Patient Matching Matters

Patient matching links every record to the right person. A duplicate occurs when one person has two records, splitting their history. An overlay occurs when two people's information is merged into one record, which is more dangerous because one patient's allergies, blood type or diagnoses may appear in another's chart. Overlaps occur when a person has different identifiers at different facilities. Kettle River's two hospitals and clinics used separate registration systems until five years ago, leaving many overlaps.

Causes of Matching Errors

Most matching errors start at registration. Staff search under a misspelled name, a maiden name or an old address, find nothing and create a new record. Hyphenated and compound surnames, nicknames, twins and patients without Social Security numbers make searches harder. The emergency department, where patients may arrive unable to give details, creates the most temporary records that later need merging.

What Standardization Can and Cannot Do

One popular fix is standardizing demographic fields, such as formatting addresses and names consistently. Grannis et al. (2019) tested this with several real-world datasets and found that standardizing addresses improved matching sensitivity modestly, by roughly half a percentage point in one dataset and about four and a half in another, but overall accuracy did not change because specificity fell. Standardization helps, but it is not enough on its own. Kettle River will combine standardized registration fields with better search training, photo capture at registration and regular steward review of possible duplicates.

Governing Demographic Data

Race and ethnicity are missing for 19% of Kettle River patients, most often in the emergency department. The same research by Polubriaginof et al. (2019) that documented widespread gaps also showed that most patients, 86%, give usable answers when they fill in the question themselves. The demographics steward will propose a single script and category list for every site and allow patients to self-report through the portal and check-in tablets, with registration staff trained to explain why the question matters.

Policies the Committee Should Adopt First

Four policies would have the greatest early effect: a registration search standard requiring at least three search attempts with name variations before a new record is created; a demographic data standard defining formats and allowed values; a data definition policy requiring every board-reported measure to have a dictionary entry; and a data access policy defining who may view identifiable data in the warehouse.

The Cost of Doing Nothing

Poor matching carries costs that rarely appear in a budget. Each confirmed duplicate takes the MPI coordinator about 20 minutes to research and merge, and the backlog of flagged pairs now exceeds 9,000. Split records force clinicians to repeat tests when earlier results sit in the other chart, and exchange partners receive incomplete histories. An overlay can require weeks of work to separate two people's information and may need to be reported as a privacy incident. Prevention at registration is far cheaper than cleanup.

Risks and Culture

Governance fails when it becomes a paperwork exercise. To avoid that, the committee will focus on a few visible problems in its first year, publish results and recognize stewards' work. Registration staff, whose workflow changes most, will be involved in writing the search standard rather than simply receiving it.

Conclusion

Clear ownership turns data problems from everyone's complaint into someone's responsibility. A committee with authority, owners and stewards with defined decision rights and quarterly quality scorecards give Kettle River the structure to reduce duplicates, standardize demographic data and keep definitions consistent. Project Two will build these into a full data quality improvement plan.

References

Grannis, S. J., Xu, H., Vest, J. R., Kasthurirathne, S., Bo, N., Moscovitch, B., Torkzadeh, R., & Rising, J. (2019). Evaluating the effect of data standardization and validation on patient matching accuracy. Journal of the American Medical Informatics Association, 26(5), 447-456. https://doi.org/10.1093/jamia/ocy191

Kahn, M. G., Callahan, T. J., Barnard, J., Bauck, A. E., Brown, J., Davidson, B. N., Estiri, H., Goerg, C., Holve, E., Johnson, S. G., Liaw, S.-T., Hamilton-Lopez, M., Meeker, D., Ong, T. C., Ryan, P., Shang, N., Weiskopf, N. G., Weng, C., Zozus, M. N., & Schilling, L. (2016). A harmonized data quality assessment terminology and framework for the secondary use of electronic health record data. eGEMs, 4(1), Article 18. https://doi.org/10.13063/2327-9214.1244

Polubriaginof, F. C. G., Ryan, P., Salmasian, H., Shapiro, A. W., Perotte, A., Safford, M. M., Hripcsak, G., Smith, S., Tatonetti, N. P., & Vawdrey, D. K. (2019). Challenges with quality of race and ethnicity data in observational databases. Journal of the American Medical Informatics Association, 26(8-9), 730-736. https://doi.org/10.1093/jamia/ocz113