BACKGROUND: High-quality evidence indicates that intervening with health care providers improves human papillomavirus (HPV) vaccine delivery. However, scaling up evidence-based strategies in real-world clinical practice remains challenging. We sought to improve the reach and impact of strategies for HPV vaccination quality improvement (QI) through local adaptation and implementation in a large, not-for-profit health care system.
METHODS: We conducted an HPV vaccination QI program using existing materials to support physician training coupled with assessment and feedback. Local physicians with high HPV vaccination rates facilitated training, which included didactic instruction and video vignettes modeling effective communication. We randomly assigned 25 clinics with 77 physicians to the QI arm or the wait-list control arm. We used hierarchical linear models to assess HPV vaccination coverage (=1 dose) over 6 months among patients aged 12 to 14.
RESULTS: Of 45 physicians in the QI arm, the program reached 43 (95%) with training plus assessment and feedback. In the overall sample, HPV vaccination coverage increased in both the QI and control arms (8.6 vs 6.4 percentage points, respectively), although the 2.2-percentage point difference did not reach statistical significance. Sensitivity analyses that excluded physicians with poor data quality indicated a statistically significant advantage of 3.3 percentage points for QI versus control (b = 0.034; SE = 0.015; P < .05).
CONCLUSIONS: Our locally adapted QI program achieved excellent reach, with small improvements in HPV vaccination coverage. Future implementation research is needed to bolster program impact and support health systems in leveraging local resources to conduct these programs efficiently.
This is an interesting QI project examining physician education and feedback to increase HPV vaccination rates. The strength of the study include the fact that the investigators randomized the clinics (but not physicians) to either the QI intervention or control. The intervention included a 1-hour education session followed by feedback of HPV vaccination coverage at 3 time points. Overall, no statistically significant increase was seen in an intention-to-treat analysis, but after eliminating six physicians with questionable data quality, the QI group achieved ~3% higher vaccination rates. This study reinforces the difficulty in improving quality care measures, and how education and feedback are typically insufficient to attain substantial improvement.