| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the Education Development Center, Inc. (EDC), Newton; Harvard Pilgrim Health Care Plan, Boston, Massachusetts; and Dartmouth Medical School, Hanover, New Hampshire.
Address reprint requests to: Lydia O’Donnell, EdD Education Development Center, Inc. 55 Chapel Street Newton, MA 02458-1060 E-mail: lodonnell{at}edc.org
 |
Abstract |
|---|
Top
 Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Methods: We developed a comprehensive office systems intervention to facilitate screening and follow-up for women exposed to DES in utero, consisting of a DES toolkit and the clinical and administrative education necessary to use the tools effectively. The intervention was implemented in the internal medicine and obstetrics-gynecology departments at six free-standing health centers in a Boston-area staff-model health maintenance organization. Intervention sites were matched and paired with a comparison group of centers. Intervention effectiveness was assessed through pretest and posttest surveys of clinicians, medical record review of 3900 women, and review of a computerized medical records data base.
Results: There was significantly higher DES awareness and knowledge among clinical staff at intervention sites. Documentation of DES exposure in the medical record ranged from 1.14 to 2.31 times greater at intervention sites than in matched comparison sites, and rates of DES code use in pregnancy were 1.91 to 3.61 times greater.
Conclusions: The office systems intervention improved documentation of DES exposure in a managed care environment. Because this approach was designed to accommodate the limited time allotted for each patient visit, it not only improved DES screening but could also serve as a model for integrating screening for other low-prevalence but potentially serious conditions into routine care.
In a health-care system increasingly constrained by precepts of managed care, clinicians might struggle to find time for appropriate screening and prevention services in routine patient care.1,2 Screening for rare conditions can be especially difficult. Many low-incidence conditions have potentially high rates of morbidity and mortality, so systems for screening and treatment that do not add unrealistic burdens to care-givers are needed. We report the effectiveness of an office management system intervention for improving identification of women exposed in utero to diethylstilbestrol (DES), the synthetic estrogen given to an estimated 500,000 to 2 million pregnant women in the United States between 1938 and 1971.3
Diethylstilbestrol exposure in utero was first associated with vaginal clear cell carcinoma in young women in 1971,4,5 and it was subsequently related to increased frequency of vaginal adenosis and squamous metaplasia,6 unfavorable pregnancy outcomes,7 and impaired fertility. Clear cell adenocarcinoma recently has been diagnosed in exposed women up to 48 years old, and there are concerns about still unknown consequences as that cohort approaches menopause, when risks associated with tumors related to hormone changes increase.8 However, because physicians might be unaware of the need to screen DES-exposed women beyond their early 20s, the range of potential adverse effects, and the guidelines for appropriate screening tests, many providers do not ask routinely about DES exposure. Thus, health effects related to DES serve as a paradigm for other low-incidence, potentially high-risk conditions for which practice is inconsistent with recommended guidelines and for which office management system intervention might be especially appropriate.
|
Materials and Methods |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
The intervention was implemented at six centers of the health maintenance organization that were affiliated with one major teaching hospital. Those sites were matched with a comparison group of centers affiliated with a comparable teaching hospital. Intervention and comparison centers were paired by size and urban-suburban characteristics of their patient populations. At intervention sites, training was conducted during hour-long regular staff meetings for the internal medicine and obstetrics-gynecology departments. Following a structured outline, clinical leaders showed the video, presented current findings and updated guidelines for DES screening and care,10 described the toolkit, and offered feedback on current DES screening performance in the form of data collected from a survey of health maintenance organization members. Participants received folders of materials, including a letter of endorsement from the Massachusetts Medical Society, and a copy of the revised guidelines. A companion program for office staff was implemented during departmental support staff meetings.
To provide a uniform indicator of DES exposure status (positive, negative, or unknown), new computer codes were developed and included on the obstetrics-gynecology encounter forms used by intervention and comparison sites. The codes could also be entered manually into the standard internal medicine encounter form. The following three sources of data were used to assess intervention effectiveness: a pretest-posttest of clinicians’ DES-related knowledge and awareness, medical record review of a random sample of women between ages 22 and 51 years who had clinical encounters with internal medicine or obstetrics-gynecology staff, and pretest-posttest comparisons of the frequency with which DES-related diagnostic and treatment codes were used.
Pretest and posttest surveys were administered at departmental meetings to all physicians and nurse practitioners in obstetrics-gynecology and internal medicine at intervention and comparison sites. A study coordinator distributed and collected surveys and contacted and surveyed staff who had not attended meetings. The 57-item questionnaire assessed awareness of DES-related conditions, guidelines for care and referral of exposed women, and awareness of the availability of medical record codes. Attitudinal items documented clinicians’ perceptions whether exposure to DES is still a medical concern and whether women should be aware of their exposure status.
Medical record reviews were done at baseline, immediately after the intervention (4 months), and 1 year after intervention. At each round, we drew from each intervention and control site the medical records of 100 randomly selected women (ages 22–51 years) who had internal medicine or obstetrics-gynecology visits during the relevant time. Randomization was achieved by using a random number function to produce groups of integers that were matched with sequences of consecutive integers in the medical records of eligible patients. Records were reviewed by a trained research assistant using a chart review form developed and tested for reliability by three investigators who reviewed a random sample of 40 charts. Interrater reliability was greater than 80% on all items. Abstracted data included the presence or absence of any mention of DES in text or any use of DES diagnostic codes.
Clinicians at intervention and comparison sites were told of the availability of new computer codes and given revised encounter forms that included prompts for DES exposure. Using a system-wide computerized database, a monthly summary of all DES-related code use (positive, negative, or unknown) was collected according to treatment condition for 1 year, beginning the first month of intervention implementation. During the same interval, data also were collected by site on the use of codes associated with newly diagnosed pregnancy.
The pretest clinician survey was completed by 355 respondents (response rate 83%); and 254 completed the posttest (response rate 71%). The number of respondents for the posttest was lower because of lower response rate and decreases in staffing. Survey data allowed for comparison of pretest to posttest changes on measures of DES awareness and knowledge by condition. Individual items were examined to assess for consistency with which changes were greater among intervention versus control sites, then summative scales were created for indicators of DES awareness and knowledge. The awareness scale consisted of six items, and the knowledge scale was a summative score of the number of correct answers to eight questions. Each scale was tested for internal validity using a Cronbach alpha greater than .68 as an acceptable level of internal consistency. The significance of differences in pretest to posttest scale scores between matched pairs of sites were assessed using a Wilcoxon signed rank test.
Medical record data were analyzed for DES screening, use of computer codes, or text mention of exposure status. Proportionate changes were compared in use of DES indices by treatment condition and site within matched pairs. The relative magnitude of change in proportion of charts with DES indicators was assessed for each matched pair. Pairs in which change was greater for the intervention site were assigned positive scores and those with greater change for the comparison site were assigned negative scores. Scores were analyzed using a Wilcoxon signed rank test. Information from the computerized database was used for graphical comparison of cumulative frequency of DES code use over 12 months. The significance of the cumulative difference in rates of pregnancy-associated code use across matched pairs was assessed by a Wilcoxon signed rank test of the proportion of pairs in which high-intensity sites demonstrated greater use. All tests of statistical significance were two tailed.
|
Results |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Table 1
presents pretest and posttest data on indicators of clinician DES awareness and knowledge according to treatment condition. Within treatment condition, that permitted comparison of pretest to posttest differences and allowed comparison across conditions at both times. At pretest, levels of awareness were not substantially different between comparison and intervention sites; however, intervention sites showed greater increases in awareness indicators between pretest and posttest. Table 1 provides similar information on DES knowledge. Although there were pretest differences on several items between comparison and intervention sites, a pretest-posttest increase in knowledge was greater for all items at the intervention sites.
|
shows pretest-posttest differences in scale scores for each matched pair. For each scale, the difference was greater for the intervention site in every pair. Paired analysis showed that those differences were significant for both scales (P < .02, Wilcoxon signed rank test).
|
shows the relative pretest-posttest change in proportion of charts with indication of DES exposure status for each matched pair. In every pair, the change was greater for the intervention site (P < .02, Wilcoxon signed rank test). Figure 1 shows the cumulative change in code use by site and shows that directly after intervention there was a rapid increase in code use at intervention sites. Level of code use continued to increase at intervention sites throughout the 12-month follow-up period. There was a markedly less rapid increase at comparison sites. Table 4 shows that, in a paired analysis, rates of code use in pregnancy diagnoses were consistently greater in intervention sites (P < .02, Wilcoxon signed rank test).
|
|
|
|
Discussion |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
The number of women screened remains fairly low relative to the patient population. Users must weigh the magnitude of effect against the improbability that significant screening will occur without such an intervention and the potential life-saving benefit for every additional woman identified as exposed. Although intervention sites responded to the program with an immediate, sharp increase in screening, comparison sites eventually caught up in screening rates. That convergence might reflect the importance of the modified form that prompted clinicians to code for exposure. The creation of a condition-specific code and the distribution of modified forms might be the most cost-effective method to increase screening.
This program offers a useful model for rapidly increasing screening for uncommon conditions. In an atmosphere of competing priorities, a successful intervention must motivate individual clinicians to accord priority to the desired change and minimize the demand on the scarcest resource (ie, clinician time) by using other supports within the system. As managed care systems attempt to respond to many emerging quality control and risk management concerns, programs that target the office environment for practice changes can provide marked improvement to traditional methods of information diffusion.
|
Footnotes |
|---|
Received July 6, 1999. Received in revised form March 31, 2000. Accepted April 20, 2000.
|
References |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
2. Lewis CE. Disease prevention and health promotion practices of primary care physicians in the United States. Am J Prev Med 1988; 4(Suppl):9–16.[Medline]
3. Noller KL, Fish CR. Diethylstilbestrol usage: Its interesting past, important present, and questionable future. Med Clin North Am 1974;58:739–810.
4. Greenwald P, Barlow JJ, Nasca PC, Burnett WS. Vaginal cancer after maternal treatment with synthetic estrogens. N Engl J Med 1971;285:390–2.
5. Herbst AL, Uelder H, Poskanzer DC. Adenocarcinoma of the vagina: Association of maternal stilbestrol therapy with tumor appearance in young women. N Engl J Med 1971;284:878–81.
6. Robboy SJ, Szyfelbein WM, Goellner JR, Kaufman RH, Taft PD, Richard RM. Dysplasia and cytologic findings in 4,589 young women enrolled in diethylstilbestrol-adenosis (DESAD) project. Am J Obstet Gynecol 1981;140:579–86.[Medline]
7. Barnes AB, Colton T, Gundersen J, Noller KL, Tilley BC, Strama T, et al. Fertility and outcome of pregnancy in women exposed in utero to diethylstilbestrol. N Engl J Med 1980;302:609–13.[Abstract]
8. Registry for Research on Hormonal Transplacental Carcinogenesis. Clear cell adenocarcinoma collaborative studies, University of Chicago DES Program. http://obgyn.bsd.uchicago.edu/obgyn/registry.html. Accessed June 2, 2000.
9. Pommerenke FA, Dietrich A. Improving and maintaining preventive services. Part 1: Applying the patient path model. J Fam Pract 1992;34:86–91.[Medline]
10. Harvard Community Health Plan. Recommendations for treatment of women exposed to DES in utero. (unpublished guidelines), 1995.
11. Davis DA, Thomson MA, Oxman AD, Haynes RB. Changing physician performance: A systematic review of the effect of continuing medical education strategies. JAMA 1995;274:700–5.[Abstract]
12. Dietrich AJ, O’Connor GT, Keller A, Carney PA, Levy D, Whaley FS. Cancer: Improving early detection and prevention. A community practice randomized trial. BMJ 1992;304:687–91.
13. Carney PA, Dietrich AJ, Keller A, Landgraf J, O’Connor GT. Tools, teamwork, and tenacity: An office system for cancer prevention. J Fam Pract 1992;35:388–94.[Medline]
14. Davis D, O’Brien MA, Freemantle N, Wolf FM, Mazmanian P, Taylor-Vaisey A. Impact of formal continuing medical education: Do conferences, workshops, rounds and other traditional continuing education activities change physician behavior or health care outcomes? JAMA 1999;282:867–74.
15. Cohen SJ, Halvorson HW, Gosselink CA. Changing physician behavior to improve disease prevention. Prev Med 1994;23:284–91.[Medline]
16. Dietrich AJ, Barrett J, Levy D, Carney-Gerston P. Impact of an educational program on physician cancer control knowledge and activities. Am J Prev Med 1990;6:346–52.[Medline]
17. Maly RC, Abrahamsen AF, Hirsch SH, Frank JC, Reuben DB. What influences physician practice behavior? An interview study of physicians who received consultative geriatric assessment recommendations. Arch Fam Med 1996;5:448–54.[Abstract]
18. Green LW, Eriksen MP, Schor EL. Preventive practices by physicians: Behavioral determinants and potential interventions. Am J Prev Med 1988;4(Suppl):101–7.[Medline]
This article has been cited by other articles:
|
|
 |
|
  A. J. Dietrich, T. E. Oxman, J. W. Williams Jr, K. Kroenke, H. C. Schulberg, M. Bruce, and S. L. Barry Going to Scale: Re-Engineering Systems for Primary Care Treatment of Depression Ann. Fam. Med, July 1, 2004; 2(4): 301 - 304. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
–) intervention; (–•–) control.