MMHA 6900 Week 5 Postoperative VTE Analysis Example

Validate Before You Explain: A Work-System Safety Analysis of Flagged Postoperative Blood Clots

Student Name

Walden University

MMHA 6900: Healthcare Quality Management

Instructor Name

Month Day, Year

Validate Before You Explain: A Work-System Safety Analysis of Flagged Postoperative Blood Clots

The Signal

Cobalt Ridge Medical Center, the composite hospital this course follows, received a quarterly report showing that its rate of postoperative pulmonary embolism or deep vein thrombosis, as flagged by a claims-based patient safety indicator, had risen to the top quartile among peer hospitals. The surgical quality committee asked for an analysis. The tempting response is to assume that prophylaxis is being missed and to launch an education campaign. This analysis takes a different first step: it validates the flagged cases, then examines the work system around those that are real, through the lens of the SEIPS work-system model.

Why Validate First

Claims-based venous thromboembolism measures count clots that are coded in discharge data. A hospital that looks harder for clots, for example by ordering more ultrasound scans, will find and code more of them, including small clots that would never have caused symptoms. A study of hospitals in a national surgical quality program found that hospitals with higher rates of imaging for clots had higher reported clot rates, and that the measure was more strongly associated with how often hospitals looked than with how well they prevented clots (Bilimoria et al., 2013). This is surveillance bias, and it means a rising rate can reflect more vigilance rather than worse care.

The analysis therefore reviewed all 31 flagged cases from the past two quarters, an illustrative number, against the medical record.

Screening ultrasound had increased the previous year, after orthopedic surgeons began ordering routine scans before discharge for some patients. That change alone explains much of the rise in the indicator. The 14 symptomatic cases, however, are real events and deserve analysis.

Table 1

Validation of Flagged Cases (Illustrative, Two Quarters)

Fewer than half the flagged cases were symptomatic postoperative clots. Eight were not postoperative clots at all.

The Work System Around the Real Cases

Carayon et al. (2006) built SEIPS on the idea that harm grows out of how a whole work system fits together, so the model asks about the individuals involved, what they were asked to do, the equipment and software they used, the physical setting and the organization around them, and how all of that shaped the care given. Its later version adds the patient and family as active participants and describes how work systems adapt over time (Holden et al., 2013). The analysis reviewed the 14 symptomatic cases through each component.

The findings point to a system that leaves risk assessment and dose timing to individual memory and divides responsibility between teams without naming an owner. No single person failed. The work system made omissions easy and their detection unlikely.

Table 2

Work-System Findings in 14 Symptomatic Cases

Cases often had more than one finding. The most frequent was an optional risk assessment with no prompt, which let high-risk patients go unidentified.

What the Case Reviews Looked Like

Two cases illustrate the pattern. In one, a 71-year-old woman with a prior clot had a hip replacement on a Friday. The preoperative visit did not record her history because the risk assessment screen was optional and the clinic was running late. Her prophylaxis order was written for the following morning, and the weekend nurse could not find compression devices on the unit. She developed a symptomatic clot on day four. In another, a man recovering from bowel surgery was comanaged by the surgical team and a hospitalist. Each assumed the other had ordered prophylaxis, and neither had until the second postoperative day. Neither case involved a single careless act. Both followed from how the work system was arranged: optional steps, divided ownership and equipment gaps that appeared only on weekends.

These reviews also showed what worked. In most surgical cases without clots, the order set had been completed in full by a resident who used it every day. Reliability came from routine use of a well-designed tool, which is the lesson the recommendations build on.

Recommendations

The analysis recommends four changes. First, make the risk assessment a required step in postoperative order sets, with prophylaxis defaulted according to risk. Second, set a standard that the first prophylaxis dose is given within 24 hours of surgery unless a surgeon documents a bleeding concern, and report compliance monthly. Third, assign prophylaxis ownership to the surgical team for the first 48 hours, with hospitalist comanagement after that, written into the comanagement agreement. Fourth, stock compression devices on every surgical unit with a weekend par level.

Separately, the committee should report the indicator with a note on screening practices, and track symptomatic clots confirmed by chart review as an internal measure, since the claims indicator will continue to rise with surveillance.

Limits

Two quarters of cases is a small sample. Chart review depends on documentation, which may miss symptoms. The analysis did not compare Cobalt Ridge's prophylaxis rates with peers. The surveillance finding does not mean screening is wrong; it means the indicator must be read with screening practice in view.

Conclusion

The rise in Cobalt Ridge's postoperative clot indicator was partly a measurement effect: more screening found more clots, and some flags were not postoperative clots at all. Validating the cases first prevented an education campaign aimed at the wrong problem. The real cases revealed a work system with an optional risk assessment, delayed first doses, unclear ownership and missing equipment, each of which can be fixed by design rather than by exhortation.

References

Bilimoria, K. Y., Chung, J., Ju, M. H., Haut, E. R., Bentrem, D. J., Ko, C. Y., & Baker, D. W. (2013). Evaluation of surveillance bias and the validity of the venous thromboembolism quality measure. JAMA, 310(14), 1482-1489. https://doi.org/10.1001/jama.2013.280048

Carayon, P., Schoofs Hundt, A., Karsh, B. T., Gurses, A. P., Alvarado, C. J., Smith, M., & Flatley Brennan, P. (2006). Work system design for patient safety: The SEIPS model. Quality and Safety in Health Care, 15(Suppl 1), i50-i58. https://doi.org/10.1136/qshc.2005.015842

Holden, R. J., Carayon, P., Gurses, A. P., Hoonakker, P., Hundt, A. S., Ozok, A. A., & Rivera-Rodriguez, A. J. (2013). SEIPS 2.0: A human factors framework for studying and improving the work of healthcare professionals and patients. Ergonomics, 56(11), 1669-1686. https://doi.org/10.1080/00140139.2013.838643