Hostname: page-component-745bb68f8f-lrblm Total loading time: 0 Render date: 2025-02-11T12:07:41.274Z Has data issue: false hasContentIssue false
  • English
  • Français

EFFECT OF A NATIONWIDE PROGRAM OF EDUCATIONAL OUTREACH VISITS TO IMPROVE THE PROCESSES OF CARE FOR PATIENTS WITH TYPE 2 DIABETES

Published online by Cambridge University Press:  21 April 2004

Philippe Ricordeau ,
Pierre Durieux ,
Alain Weill ,
Gilles Chatellier ,
Nathalie Vallier ,
Alvine Bissery ,
Pierre Fender  and
Hubert Allemand
Philippe Ricordeau
Affiliation:
Caisse nationale d'assurance maladie des travailleurs salariés
Pierre Durieux
Affiliation:
Hôpital Européen Georges Pompidou
Alain Weill
Affiliation:
Caisse nationale d'assurance maladie des travailleurs salariés
Gilles Chatellier
Affiliation:
Hôpital Européen Georges Pompidou
Nathalie Vallier
Affiliation:
Caisse nationale d'assurance maladie des travailleurs salariés
Alvine Bissery
Affiliation:
Hôpital Européen Georges Pompidou
Pierre Fender
Affiliation:
Caisse nationale d'assurance maladie des travailleurs salariés
Hubert Allemand
Affiliation:
Caisse nationale d'assurance maladie des travailleurs salariés
Rights & Permissions [Opens in a new window]

Abstract

Objectives: To improve processes of ambulatory care for patients with type 2 diabetes in a nationwide program.

Methods: Interrupted time-series analysis with audits of practice. To implement selected recommendations of national guidelines, educational outreach visits (office visits or phone discussions) were offered to all French physicians who diagnosed one case of type 2 diabetes during a six-month intervention period. Outcome measures were the number of HBA1c measurements recorded monthly in the medical insurance computer database and the proportion of diabetic patients for whom one test had been reimbursed during the previous six months (HBA1c, fasting blood glucose) or previous twelve months (serum cholesterol, serum creatinine, urine microalbumin, electrocardiogram, ophthalmologic examination).

Results: A total of 15,522 office visits and 9,062 telephone discussions were performed among 22,940 physicians. The increase in the monthly proportion of the number of HBA1c tests to the total number of laboratory tests was higher during the intervention period than during the preintervention (p value<.0001) and postintervention periods (p value<.001). Between the first audit (n=651,574) and the third audit (n=911,871), HBA1c measurements increased from 41.2% to 60.5% and blood glucose measurements performed alone decreased from 38.8% to 18.7%. Urine microalbumin measurements increased from 10.6% before to 15.3% after intervention. Only a slight increase was observed for other tests.

Conclusions: Physician to physician outreach visits can be an effective way to improve the processes of care for diabetes and to routinize nationwide use of practice guidelines.

Keywords

Educational outreach visitsDiabetesPractice guidelinesIntervention studiesEvidence based health policy
Type
RESEARCH NOTES
Information
International Journal of Technology Assessment in Health Care , Volume 19 , Issue 4 , December 2003 , pp. 705 - 710
Copyright
© 2004 Cambridge University Press

Given the economic constraints of health-care systems, health policy makers and payers are looking for instruments, such as clinical guidelines, that can help them to justify the use of resources (6). However, there is still considerable uncertainty whether clinical guidelines will improve clinical practice and patient outcomes. Although it has been shown in small controlled environments that clinical practice guidelines can improve the quality of care (3), whether they can achieve this improvement in daily practice is unclear. These strategies have never been tested on a national population of physicians engaged in routine primary care who had not volunteered to participate in an experimental study.

In 1998, The Caisse nationale d'assurance maladie des travailleurs salariés (CNAMTS), France's largest health insurance fund developed an initiative aimed at improving physicians' medical practices by encouraging through a validated implementation strategy, educational outreach visits (10), their compliance with national guidelines on the management of type 2 diabetes (1;2). In recent years, diabetes mellitus has become a major health problem in all countries, and it has been shown that the medical management of patients with this illness is frequently inappropriate. The objective of this study was to determine whether educational outreach visits to all French physicians caring for patients with type 2 diabetes would increase the frequency with which appropriate laboratory tests were ordered for their patients and to test the effectiveness of this approach in a routinely implemented national health program.

METHODS

Setting and Study Population

The study was conducted in physicians' offices throughout France by the CNAMTS in cooperation with both the agricultural (Mutualité sociale agricole, MSA) and self-employed insurance funds (Assurance maladie des professions indépendantes, AMPI). These funds cover 95.9% of the French population.

Among the 172,500 physicians in France, a total of 64,328 physicians were considered most likely to manage type 2 diabetes mellitus: 60,580 general practitioners, and 3,748 specialists in internal medicine or endocrinology, working either in private practice or in hospitals.

Intervention

The general approach was to communicate the French guidelines regarding type 2 diabetes mellitus care during educational outreach visits performed by médecins conseils (full-time physicians salaried by health insurance funds who are responsible for monitoring the medical care delivered to patients by health professionals).

First, a booklet containing a summary of the French practice guidelines on the management of type 2 diabetes (1;2) was mailed to all French physicians in May of 1999. Then, between June of 1999 and December of 1999, a program of physician to physician outreach visits was organized to promote these guidelines. Two types of educational outreach visits were conducted: a formal outreach visit by a médecin-conseil in the practice setting of the physician or a telephone discussion between the médecin-conseil and the practicing physician (when it was difficult to schedule a formal visit, that is, in some rural areas).

During the intervention period, each time a practicing physician addressed to the CNAMTS a request for a copayment exemption for a diabetic patient (out-of-pocket copayments by patients are waived for 30 specific disorders, including diabetes mellitus), the médecin-conseil called the physician to make an appointment. During their meeting, whether in person or on the phone, both physicians discussed the type 2 diabetes guidelines and their relevancy to the particular patient for whom the waiver request had been made.

Médecins-conseils who were to participate in the outreach program were trained during a one-day regional training session (for the CNAMTS, n=1,264). The training session was based on a nationally established guide to be used by the médecin conseil during each outreach visit. The guide included recommendations on management of diabetes (glycosylated hemoglobin determination, the recommended method for monitoring type 2 diabetes, should be performed every three to four months; an electrocardiogram, a funduscopic examination of the eye, serum creatinine, serum cholesterol, and urine microalbumin determinations should be performed annually).

Outcome Measures

First, we performed an interrupted time series analysis of the monthly proportion of the number of HBA1c measurements to the total number of laboratory tests recorded in the database from January of 1998 to December of 2000 (seventeen months before, six months during, and twelve months after intervention). The National Medical Insurance's Computerized System stores the records of all insured individuals. Since 1997, biologists and pharmacists have been required to routinely code each reimbursable laboratory test or delivered drug. In 1998, coding for laboratory tests and filled prescriptions was 91.3% and 70.3% complete, respectively, and reached 97.9% and 89.0%, respectively, in 2000. The completeness with which coding was reported has been attributed to regional differences. The variation was unrelated to patient characteristics (9).

Second, three practice audits were performed among all type 2 diabetic patients: in 1999 (before), 2000 (during), and 2001 (after) the intervention. Type 2 diabetic patients were defined as any patient receiving at least one oral hypoglycemic drug. All type 2 diabetic patients recorded in the database during the first three months of the year were included in the study. The main end point was the proportion of patients for whom at least one HBA1c measurement had been reimbursed in the previous six months. The other end points were, the proportion of patients reimbursed at least once during the previous six months for fasting blood glucose and the proportion of patients reimbursed at least once during the previous year for any of the following tests: serum cholesterol and creatinine, urine microalbumin, electrocardiogram, ophthalmologic examination.

Statistical Analysis

We applied a repeated analysis of covariance with potentially structured residuals, with interventions as fixed effects, and time as linear fixed effect. We tested the effect of interventions by comparing the means of proportions of HBA1c tests, adjusted to the slope of the three periods regression line. No correlation was found between residuals. Statistical analysis was performed using Stata 7.0 (StataCorp.2001, Stata statistical Software, release 7.0, Stata Corporation, College Station, TX). Considering the number of subjects involved (the entire population of French diabetic patients), no statistical test was performed for the three audits.

RESULTS

During the intervention period, a total of 24,584 visits were performed among 22,940 physicians (15,478 physicians received a formal visit and 7,462 a phone discussion.) Among the 48,804 requests for patient waivers from copayments for type 2 diabetes sent by practicing physicians to the CNAMTS (data concerning AMPI and MSA not available), 27,612 requests were not followed by a visit: 209 visits were refused by physicians and 27,403 requests concerned physicians who had already received one visit concerning another diabetic patient during the intervention period.

An increase in the monthly proportion of the number of HBA1c tests to the total number of laboratory tests was observed during the three-year study period (Figure 1). This trend was significantly accelerated during the intervention period when physician to physician outreach visits were performed, in comparison with the preintervention (p value<.0001) and the postintervention period (p value<.001). The trend during the postintervention period did not differ significantly from that of the preintervention period (p value=.966).

Interrupted time series analysis.

Between the first and the third audit, we observed a 46.8% increase in HBA1c measurements (from 41.2% before to 60.5% after intervention) along with a 51.8% decrease in blood glucose measurements performed alone (from 38.8% before to 18.7% after intervention). Urine microalbumin detection increased by 44.3% (from 10.6% before to 15.3% after intervention). Only a slight increase was observed for other tests (Table 1).

DISCUSSION

The implementation of clinical guidelines for type 2 diabetes using outreach visits resulted in an increase in the number of HBA1c tests ordered during the intervention period. These results are consistent with previous studies (8), but there is still considerable room for improvement. Despite this large-scale intervention, no HBA1c test was reimbursed for almost 40% of patients with type 2 diabetes. The impact of the intervention on other tests was very small. To achieve a long-term impact on processes of care, physician to physician outreach visits should be repeated or other change methods are needed to supplement the visits.

This study has two main strengths: First, it involved 22,940 practicing physicians (35.7% of the physicians who were considered more likely to manage diabetic patients) and less than 1% of the physicians contacted refused to participate. Thus, this intervention could be an effective way to implement national guidelines in a nationwide program among physicians involved in routine patient care. Our study has shown that evidence-based health policies are feasible (4).

Second, all the data presented comes from the computerized information system of local CNAMTS agencies. Assessing quality by using administrative data has been questioned (5). Our study shows that well-maintained large pharmacy and laboratory-test databases contain simple process-based indicators that can be used to monitor health-care quality programs (7).

However, this study has some limits: First, we have no data comparing the impact on processes of care for patients followed by physicians who had an office visit versus those who had a phone discussion or who were not visited. Practicing physicians were not invited to participate in the study before its onset. We considered the confidentiality of information to be an important reason for the high acceptance of the program. Other limits of this study are related to the specificities of the CNAMTS information system: Data used for the interrupted time series analysis were not limited to patients with diabetes but concerned all patients waived for copayment. However, because HBA1c determinations are only useful for diabetes management, it is unlikely that it was prescribed for the management of another disease. Audits of practice concerned only diabetic patients taking drugs. Patients treated only with diet and exercise are not included in the data base. In addition, data used to monitor the intervention's impact did not allow us to evaluate other important aspects of diabetes management emphasized in the guideline, such as blood pressure measurement, foot examinations, or dietary advice.

These limitations are mainly due to the fact that our program was intended to be both nationwide and included in the daily practice of physicians. Additional studies are needed, in particular at a local level, to evaluate the amount of reinforcement required to ensure the stability of changes in physician behavior and to evaluate its impact on other aspects of diabetes management.

POLICY IMPLICATIONS

(1) Outreach visits can be an effective way to implement national guidelines in a nationwide program among physicians involved in routine patient care, but to achieve a long-term impact on processes of care, physician to physician outreach visits should be repeated or other change methods are needed to supplement the visits. (2) National clinical guidelines could have an impact on processes of care when validated implementation strategies are used. (3) Health policies aimed at changing clinical practice should be based on scientific evidence. (4) Administrative databases can be used to monitor health-care quality programs.

The authors are particularly grateful to Robert Bourrel for his participation in data collection and to Donald Schwartz, MD, for his help in proofreading the English translation of the manuscript. The authors thank all the médecins-conseils, their teams, and all the practicing physicians who accepted to participate in the study. The authors received greatly appreciated assistance from the thoughtful review of Miriam Orleans, PhD (University of Colorado School of Medicine) and Michael Fine, MD (University of Pittsburgh Medical Center). This study was supported by the Caisse nationale d'assurance maladie des travailleurs salariés.

References

1999 Agence française de sécurité sanitaire des produits de santé. Traitement médicamenteux du diabète de type 2. Paris: AFSSAPS; Available at: http://afssaps.sante.fr. Accessed October 1, 2003.
2000 Agence nationale pour le développement de l'évaluation médicale. Strategy for management of type 2 diabetes excluding management of complications. Paris: ANAES; Available at: www.anaes.fr. Accessed October 1, 2003.
Bero LA, Grilli R, Grimshaw JM et al. 1998 Closing the gap between research and practice: An overview of systemic reviews of interventions to promote the implementation of research findings. BMJ. 317: 465- 468.Google Scholar
Garber AM. 2001 Evidence-based coverage policy. Health Aff (Millwood). 20: 62- 82.Google Scholar
Iezzoni LI. 1997 Assessing quality using administrative data. Ann Intern Med. 127: 666- 674.Google Scholar
Lohr KN, Eleazer K, Mauskopf J. 1998 Health policy issues and applications for evidence-based medicine and clinical practice guidelines. Health Policy. 46: 1- 19.Google Scholar
Palmer RH. 1997 Process-based measures of quality: The need for detailed clinical data in large health care databases. Ann Intern Med. 127: 733- 738.Google Scholar
Renders CM, Valk GD, Griffin S et al. 2001 Interventions to improve the management of diabetes mellitus in primary care, outpatient and community settings (Cochrane review). In: The Cochrane Library, Issue 2, Oxford: Update Software.
Ricordeau P, Weill A, Vallier N et al. 2000 L'épidémiologie du diabète en France métropolitaine. Diabetes Metab. 26(Suppl 6): 11- 24.Google Scholar
Thomson O'Brien MA, Oxman AD et al. 2000 Educational outreach visits to improve health professional practice and health care outcome. (Cochrane Review). In: The Cochrane Library, Issue 1, Oxford: Update Software.
Figure 0

Interrupted time series analysis.

Figure 1

Results of the Practice Audits