The National Nursing Home Survey (1) estimated that 14.5% of U.S. nursing-home residents (189,000 persons) were diagnosed as having diabetes mellitus in 1977. In Colorado, there are approximately 16,000 diabetic nursing-home residents. A review of 1977 and 1978 hospitalization data showed that diabetic nursing-home residents in Colorado had consistently higher proportions of hospitalizations for diabetic ketoacidosis and coma, insulin reactions, and amputations than diabetic persons aged 65 years and older not residing in long-term care facilities. Some of the increased hospital utilization by nursing-home residents may be due to concurrent illness, medication regime, and disability. However, in a setting with care and supervision provided by health professionals, complications, such as insulin reactions and acute hyperglycemia, should be largely preventable or recognized early enough to avoid hospitalization. Therefore, it was postulated that education for professional staff in nursing homes to improve care practices would be effective in reducing excess hospital utilization and the associated costs.

Written policies and procedures are one means by which standards of care within a facility can be measured. The Colorado Diabetes Control Program (DCP) adapted criteria for diabetes-care policies and procedures from the "Guidelines for Diabetes Care in Skilled Nursing Facilities," developed by the American Diabetes Association and the American Association of Diabetes Educators (2). One hundred twenty-five criteria were developed covering 10 content areas: dietary department practices, care during acute illness, foot and skin care, care of hypoglycemic reactions, activity planning, administering medication, urine testing, blood testing, diet management by nursing staff, and patient education. When appropriate, individual criteria within each content area included: patient assessment, methods for specific tasks, minimum frequency for performing tasks, reporting actions, and documentation. The Colorado DCP staff reviewed each facility's policies and procedures for the presence or absence of the 125 criteria.

In 1981, before the professional education intervention conducted by the Colorado DCP, written diabetes-care policies and procedures in 29 Denver metropolitan-area nursing homes were compared to the 125 criteria. The average institution met 40.6 (range 17-69) criteria. Areas with major deficiencies were care during acute illness, care of hypoglycemia, and foot and skin care.

Workshops and follow-up consultation were then offered to administrative nursing-home staff to assist them in developing and implementing diabetes-care policies and procedures. These consisted of a series of three workshops for directors of nursing, inservice program directors, dietitians, food-service supervisors, and administrators. The first workshop reviewed diabetes care of the elderly; the second covered policy and procedure development and focused on definitions, benefits, format, implementation, and diabetes-care recommendations; and the third covered educational principles and the development of inservice programs. In addition, staff inservice and orientation workshops were conducted. At the completion of the workshops, the DCP staff offered consultation services to all participants to assist them in developing policies and procedures, as well as educational programs. The staff also contacted the participants periodically during the following year to offer them assistance and to monitor progress.

One year after this intervention (1982), the average number of diabetes policies and procedures had increased significantly to 48.9 (p 0.01). In 1983, after repeat workshops and continued consultation, the average number of policies and procedures increased to 63.5.

In an effort to determine the effect of intervention on frequency and duration of hospitalization, hospitalization data on diabetic residents of the facilities for the 1-year period before intervention (1981) and a corresponding period 2 years later (1983) were obtained from an audit of nursing-home records, which were the primary data source for this project. Nursing homes routinely request hospitalization information when patients return after hospitalization. The information the nursing homes receive varies with each hospital. If the hospital does not send information as requested or if the patient does not return to the facility, it is difficult to obtain the patient's permission to release records, since many patients are legally incompetent, and family members are not always accessible. For this study, the nurse reviewer was instructed to determine reasons for hospitalization from the following sources: hospital admission summary, hospital discharge summary, hospital transfer to nursing home form, nursing-home progress notes, nursing-home transfer to hospital form. Laboratory values, when available, were used to classify acute hyperglycemic and hypoglycemic reactions; otherwise, physician diagnosis or symptom history was used. Other problems with using nursing-home records were limited physician progress notes and nursing-care documentation. Often, nursing activity flowsheets are not in current records.

There were 471 diabetic patients (mean age 78.0; 76%, female; 47%, using insulin; average length of nursing-home stay, 23.5 months) in the 29 facilities (33.8% skilled-nursing homes). Hospitalizations included in the analysis were limited to those for acute hyperglycemia, diabetic ketoacidosis, hyperosmolar nonketotic coma, hypoglycemia, and lower-extremity lesions with or without amputations. There were 33 such hospitalizations among 325 residents preintervention and 39 diabetes hospitalizations among 358 residents postintervention. No significant change occurred in the diabetes hospitalization rates from preintervention to postintervention (12.9 hospitalizations/100 person years to 13.8 hospitalizations/100 person years,(p 0.05). However, the hospital-days rate (the number of hospital days divided by the number of days at risk for hospitalization from nursing home) decreased significantly from 185.5 hospital days/100 person years to 133.8 hospital days/100 person years (28%) (p 0.001). In addition, the average length of stay decreased by 4.7 days from 14.4 days preintervention to 9.7 days postintervention (p = 0.06). By contrast, the average length of stay for all Colorado hospitalizations was stable during this period at 6.1 days (3).

To further evaluate the effectiveness of the intervention, the 29 nursing homes were divided into six groups according to the number of policies and procedures at baseline (low = 17-36 policies; medium = 37-44; high = 45-61) and the increase in the number of policies and procedures 1-year postintervention (low change--less than seven policies added; high change--seven or more policies added). The distribution of patients by age, sex, and level of care did not differ significantly among these six groups. The average length of hospital stay dropped consistently in all six groups following the intervention. Figure 1 shows the change in the hospital-days rate from preintervention to postintervention for the six groups. Two of the groups had patterns that were difficult to analyze; group 5 is a single, large nursing home, and group 3 had preintervention rates that were much lower than expected based on rates in all facilities. The two groups with the highest numbers of baseline policies and procedures (groups 1 and 2) had lower hospital-days rates preintervention, and these rates did not change significantly after intervention. However, groups 4 and 6 with low and medium baseline policies and procedures and a higher level of change had higher preintervention hospital-days rates which decreased significantly (p 0.001) following intervention. Reported by RF Hamman, MD, Dept of Preventive Medicine and Biometrics, University of Colorado School of Medicine, Denver, SL Michael, MS, SM Keefer, WF Young, MA, Diabetes Control Program, Colorado Dept of Health; Div of Diabetes Control, Center for Prevention Svcs, CDC.

Editorial Note

Editorial Note: It appears that intervention aimed at changing the policies and procedures of nursing homes is an effective way to decrease hospital days for diabetic residents. Furthermore, targeting the intervention at facilities with the fewest policies should be more efficient in effecting this change. Better monitoring following intervention may result in earlier referral to the hospital and admission of less acutely ill patients. This would account for a shorter length of stay but no decline in the hospital episode rate. The 28% reduction in the hospital-days rate suggests that substantial economic savings in hospital costs could occur by widespread use of this type of intervention.

Although Medicaid is the primary source of payment for most nursing-home care, most hospitalizations in this population are reimbursed by Medicare. This project was completed before the implementation of Medicare payment for hospitalization by Diagnosis Related Groups (DRGs) in October 1983. Therefore, the decrease in length of hospital stay cannot be attributed to the change in

Disclaimer   All MMWR HTML documents published before January 1993 are electronic conversions from ASCII text into HTML. This conversion may have resulted in character translation or format errors in the HTML version. Users should not rely on this HTML document, but are referred to the original MMWR paper copy for the official text, figures, and tables. An original paper copy of this issue can be obtained from the Superintendent of Documents, U.S. Government Printing Office (GPO), Washington, DC 20402-9371; telephone: (202) 512-1800. Contact GPO for current prices.

Page converted: 08/05/98

HOME  |  ABOUT MMWR  |  MMWR SEARCH  |  DOWNLOADS  |  RSS |  CONTACT POLICY  |  DISCLAIMER  |  ACCESSIBILITY

Morbidity and Mortality Weekly Report Centers for Disease Control and Prevention 1600 Clifton Rd, MailStop E-90, Atlanta, GA 30333, U.S.A Department of Health and Human Services

This page last reviewed 5/2/01