Nurse practitioner-based diabetes care management: Impact of telehealth or telephone intervention on glycemic control

Background: Limited evidence exists on the impact of nurse practitioner-managed diabetes mellitus care coordination programs in the primary care setting and specifically on the use of telehealth to manage veterans with diabetes in the home. The number of days of participation in the program was greater for the telehealth group than the group receiving the telephonic intervention (192.2 vs 161.9) but this difference was not statistically significant (p = 0.13). Approximately 75% of the patients (n = 192) worked with nurse practitioners and had reached individualized glycemic goals at disenrollment. Among these patients, those receiving the telehealth intervention had a 3.1% (SD = 1.9, p < 0.001) reduction in HbA1c and those receiving the telephone intervention had a 2.7% (SD = 1.9, p < 0.001) reduction in HbA1c, over a mean period of 204 days. Both interventions lost some of their effect following program disenrollment. The mean rise in HbA1c in the post-program period was 0.69% for the telehealth intervention and 0.63% for the telephone intervention (the average number of post-program days was 434 days for the telehealth intervention and 323 days for the telephone intervention). After adjusting for HbA1c at disenrollment and the number of days between disenrollment and the latest HbA1c measurement, no significant difference in the rise in HbA1c was seen between the two interventions (p = 0.80). Objective: To compare the impact of nurse practitioner-based diabetes mellitus care management programs using either a telehealth or a telephone intervention. Specific aims were to (i) compare the efficacy of telehealth and telephone interventions in a diabetes care management program, with regards to glycemic control; (ii) examine the impact of program exposure on the control of diabetes following patient disenrollment from the program; and (iii) identify the average duration of use of a telehealth or telephone intervention required to reach individualized glycemic goals. The number of days of participation in the program was greater for the telehealth group than the group receiving the telephonic intervention (192.2 vs 161.9) but this difference was not statistically significant (p = 0.13). Approximately 75% of the patients (n = 192) worked with nurse practitioners and had reached individualized glycemic goals at disenrollment. Among these patients, those receiving the telehealth intervention had a 3.1% (SD = 1.9, p < 0.001) reduction in HbA1c and those receiving the telephone intervention had a 2.7% (SD = 1.9, p < 0.001) reduction in HbA1c, over a mean period of 204 days. Both interventions lost some of their effect following program disenrollment. The mean rise in HbA1c in the post-program period was 0.69% for the telehealth intervention and 0.63% for the telephone intervention (the average number of post-program days was 434 days for the telehealth intervention and 323 days for the telephone intervention). After adjusting for HbA1c at disenrollment and the number of days between disenrollment and the latest HbA1c measurement, no significant difference in the rise in HbA1c was seen between the two interventions (p = 0.80). Design, setting, and patient population: A retrospective pre-post cohort study of a nurse practitioner-managed diabetes care coordination program was performed in primary care clinics in a Midwest Veterans Administration Medical Center in the US. The cohort included in this study consisted of 259 patients who were enrolled in the program between August 2003 and October 2005 and who disenrolled from the program before January 2006. The number of days of participation in the program was greater for the telehealth group than the group receiving the telephonic intervention (192.2 vs 161.9) but this difference was not statistically significant (p = 0.13). Approximately 75% of the patients (n = 192) worked with nurse practitioners and had reached individualized glycemic goals at disenrollment. Among these patients, those receiving the telehealth intervention had a 3.1% (SD = 1.9, p < 0.001) reduction in HbA1c and those receiving the telephone intervention had a 2.7% (SD = 1.9, p < 0.001) reduction in HbA1c, over a mean period of 204 days. Both interventions lost some of their effect following program disenrollment. The mean rise in HbA1c in the post-program period was 0.69% for the telehealth intervention and 0.63% for the telephone intervention (the average number of post-program days was 434 days for the telehealth intervention and 323 days for the telephone intervention). After adjusting for HbA1c at disenrollment and the number of days between disenrollment and the latest HbA1c measurement, no significant difference in the rise in HbA1c was seen between the two interventions (p = 0.80). Results: The mean reductions in glycosylated hemoglobin (HbA1c) associated with the program were 2.4% for the telehealth intervention (baseline 9.86%; end of program 7.46%) and 2.39% for telephone intervention (baseline 9.75%; end of program 7.36%). No significant difference in the reduction in HbA1c was noted between telehealth and telephone interventions (p = 0.96) after adjusting for baseline HbA1c and age. The number of days of participation in the program was greater for the telehealth group than the group receiving the telephonic intervention (192.2 vs 161.9) but this difference was not statistically significant (p = 0.13). Approximately 75% of the patients (n = 192) worked with nurse practitioners and had reached individualized glycemic goals at disenrollment. Among these patients, those receiving the telehealth intervention had a 3.1% (SD = 1.9, p < 0.001) reduction in HbA1c and those receiving the telephone intervention had a 2.7% (SD = 1.9, p < 0.001) reduction in HbA1c, over a mean period of 204 days. Both interventions lost some of their effect following program disenrollment. The mean rise in HbA1c in the post-program period was 0.69% for the telehealth intervention and 0.63% for the telephone intervention (the average number of post-program days was 434 days for the telehealth intervention and 323 days for the telephone intervention). After adjusting for HbA1c at disenrollment and the number of days between disenrollment and the latest HbA1c measurement, no significant difference in the rise in HbA1c was seen between the two interventions (p = 0.80). Conclusion: When employed for a comparable number of days, telehealth and telephone communication technologies used by nurse practitioners to provide individualized diabetes care management have similar effects on glycemic control. After disenrollment, HbA1c increased slightly, suggesting that veterans need continuous individualized care, in addition to routine follow-up, to manage their diabetes.