Can improving physician office systems lead to increased use of preventive care among young children?

Rates of preventive care increased among young children who were patients of primary care practices and clinics that collaborated with researchers to adopt and enhance quality-improvement systems, such as chart prescreening, risk-assessment forms, flowsheets, prompting and reminder systems, and patient education materials.

Why is this important?

Some research finds that primary care practitioners can improve their ability to deliver preventive care when they adopt a systematic, organized approach—such as using information tools, improving office processes, and increasing teamwork among office staff (Dickey et al. 1999; Dietrich et al. 1997; Goodwin et al. 2001). Many primary care physicians do not employ such approaches, however (Audet et al. 2005; Dickey and Kamerow 1996), and attempts to increase their use have not always proven successful (Solberg et al. 2000). In a recent national survey of pediatricians, only 16 percent reported using systems to remind parents about upcoming or overdue immunizations (Tierney et al. 2003)—a technique that has been shown to be effective for increasing immunization rates (Szilagyi et al. 2000).

Interventions

In a year-long, community-wide collaborative pilot project, all major primary care group practices and clinics serving young children in Durham, North Carolina, established multidisciplinary teams and received technical assistance to set objectives, monitor performance, and adopt or enhance quality-improvement systems to increase preventive care for young children (Bordley et al. 2001). Interventions were directed at one or more levels:

• Patients—educational materials and activation cards to prompt discussions with the doctor or nurse;
• Practitioners—post-it reminders on patient charts and risk assessments to prompt screening when appropriate; and
• Practice—chart prescreening to identify needed services, flowsheets indicating recommended age-specific services, and tracking systems to identify patients in need of care.

In a subsequent controlled trial, 44 pediatric and family practices in North Carolina were randomly selected and assigned to intervention and control groups to validate the effects of the intervention. Multidisciplinary local practice teams received data feedback to identify performance gaps, a set of tools to improve performance, and continuing medical education during the course of a year to learn, test, and apply quality-improvement methods in their practices (Margolis et al. 2004).

Findings

In the pilot project, the combined rates of preventive care provided to patients of the eight practices increased significantly for three of four project goals compared with rates before the intervention:

• The immunization rate increased by 7 percentage points at age 12 months and by 12 percentage points at age 24 months.
• Screening for anemia increased by 30 percentage points.
• Screening for lead poisoning (including a risk assessment or blood testing) increased by 36 percentage points (Bordley et al. 2001).
• Rates of screening for tuberculosis increased at practices that focused on this objective but not in aggregate (data not shown).

In the randomized controlled trial, rates of the four age-appropriate preventive services—immunizations and screening for tuberculosis, anemia, and lead poisoning by age 24 months—were comparable between intervention and control groups at baseline. Thirty months after the start of the intervention:

• The proportion of children who received all four preventive services increased by 17 percentage points in the intervention group (from 7 percent to 34 percent). In contrast, there was little change in the proportion of control group patients receiving all four services (from 9 percent to 10 percent) during this time.
• The combined rate of preventive care delivery was almost five times greater in the intervention group than in the control group, after adjusting for differences in the groups at the beginning of the study (Margolis et al. 2004).
• Rates of screening for tuberculosis, anemia, and lead poisoning were significantly higher in the intervention group compared with the control group. Immunization rates were similar in both groups (data not shown).

Implications

With assistance, primary care practitioners can establish and improve office systems to increase preventive care for children. In the pilot project, the degree of improvement varied among practices depending on whether they targeted a particular service for improvement and based on their ability to successfully implement changes in office systems. Longer time of exposure to the intervention was associated with increased rates of preventive care (Bordley et al. 2001). Likewise, the effects of the intervention increased over 30 months of follow-up in the controlled trial, suggesting the importance of providing a sustained implementation period to measure the impact of quality-improvement interventions. The researchers attributed the success of the intervention to ongoing support from quality experts and an emphasis on the use of measurement, tools, and rapid testing to build confidence and momentum for change (Margolis et al. 2004).

Improvement Ideas and Resources

Additional improvement resources are available from the Center for Children's Healthcare Improvement.

Measure:

The pilot project included patients of physician practices in Durham, N.C., that enrolled at least five patients per month, including pediatric and family practices, an HMO pediatric clinic, a university medical center, and a community health center. Immunization rates at 12 months included three doses of diphtheria-tetanus-pertussis (DTP) vaccine, two doses of oral or inactivated poliovirus vaccine (OPV/IPV), and two doses of Haemophylus influenzae type b (Hib) vaccine. The 24-month immunization rate included four doses of DTP, three doses of OPV/IPV, one dose of measles-mumps-rubella (MMR) vaccine, and three doses of Hib (4:3:1:3 series). Screening consisted of recognized laboratory testing or (for tuberculosis and lead poisoning) a risk assessment. All rates shown in the chart were significantly different at follow-up from baseline (Bordley et al. 2001).The randomized controlled trial consisted of 44 pediatric and family practices selected at random from 453 nonacademic practices located in two regions of North Carolina and that had a minimum number of newborns enrolling each month. Practices were randomly assigned to intervention and control groups within strata by practice size, number of newborns enrolling each month, and annual Medicaid billing. Immunization rates included four doses of DTP, three doses of OPV/IPV, one dose of MMR, three doses of Hib, and three doses of Hepatitis B vaccine (4:3:1:3:3 series) by 24 months. Screening consisted of recognized laboratory testing or risk assessment by 24 months. Results were analyzed using a logistic random regression model (Margolis et al. 2004).

Limitations:

The authors noted that some of the change in rates might have reflected improvement in documentation rather than an increase in service delivery, although improved documentation is also beneficial because it "facilitates efficient care, makes screening for needed services easier, prevents unnecessary repeated services (e.g., overimmunization), and improves communication between providers within a practice" (Bordley et al. 2001). Results of the randomized controlled trial are not generalizable to small physician practices.

Source:

Results were analyzed and reported by researchers at the University of North Carolina (Bordley et al. 2001; Margolis et al. 2004). For the pilot project, researchers abstracted a random sample of medical records at each practice at baseline (N=339) and at 12 months after each practice's office systems were operational from each of three age groups: 12 to 18 months (N=289), 19 to 24 months (N=285), and 25 to 30 months (N=300). For the controlled trial, researchers abstracted repeated random samples of medical records in each practice at baseline and at 12, 18, 24, and 30 months after the start of the intervention.

References:

* Indicates source of data used in the chart(s).Audet, A. M., M. M. Doty, J. Shamasdin et al. 2005. Measure, Learn, and Improve: Physicians' Involvement in Quality Improvement. Health Affairs (Millwood) 24 (3): 843–53. * Bordley, W. C., P. A. Margolis, J. Stuart et al. 2001. Improving Preventive Service Delivery Through Office Systems. Pediatrics 108 (3): E41. Dickey, L. L., D. H. Gemson, and P. Carney. 1999. Office System Interventions Supporting Primary Care-Based Health Behavior Change Counseling. American Journal of Preventive Medicine 17 (4): 299–308. Dickey, L. L., and D. B. Kamerow. 1996. Primary Care Physicians' Use of Office Resources in the Provision of Preventive Care. Archives of Family Medicine 5 (7): 399–404. Dietrich, A. J., P. A. Carney, C. W. Winchell et al. 1997. An Office Systems Approach to Cancer Prevention in Primary Care. Cancer Practice 5 (6): 375–81. Goodwin, M. A., S. J. Zyzanski, S. Zronek et al. 2001. A Clinical Trial of Tailored Office Systems for Preventive Service Delivery. The Study to Enhance Prevention by Understanding Practice (STEP-UP). American Journal of Preventive Medicine 21 (1): 20–8. * Margolis, P. A., C. M. Lannon, J. M. Stuart et al. 2004. Practice Based Education to Improve Delivery Systems for Prevention in Primary Care: Randomised Trial. BMJ 328 (7436): 388. Solberg, L. I., T. E. Kottke, M. L. Brekke et al. 2000. Failure of a Continuous Quality Improvement Intervention to Increase the Delivery of Preventive Services. A Randomized Trial. Effective Clinical Practice 3 (3): 105–15. Szilagyi, P. G., C. Bordley, J. C. Vann et al. 2000. Effect of Patient Reminder/Recall Interventions on Immunization Rates: A Review. Journal of the American Medical Association 284 (14): 1820–7. Tierney, C. D., H. Yusuf, S. R. McMahon et al. 2003. Adoption of Reminder and Recall Messages for Immunizations by Pediatricians and Public Health Clinics. Pediatrics 112 (5): 1076–82.