Level of service provided by DSNs to inpatients with diabetes: is it enough?

There is a general concern among diabetes specialist nurses (DSNs) regarding the level of service provided to patients with diabetes admitted to the acute unit. Since the development of services to the wards has resource implications, it is essential first to establish the potential inpatient workload and the level of services being provided in any particular region.

This article reports two audits undertaken in the Wessex region to determine:

1. The services provided to inpatients by DSNs within the region
2. The number of inpatients with diabetes in the authors’ hospital who may require DSN intervention. 

The results are presented simultaneously.

Methodology
Data on all inpatients with diabetes were collected according to the protocol shown in Table 1 and a proforma on one day per month for four consecutive months between 1 January and 30 April 1997. During the same period a questionnaire (Figure 1) was circulated to all DSNs (n = 38) within the Wessex region.

Results
Data were collected on 138 inpatients (age range 21–93 years; mean 71 years). Of these, 52% were over 75 years of age; the high proportion of elderly inpatients reflects local demographic trends in that 13% of patients served by the Royal West Sussex Trust are aged 75 years or more.

Of the 138 patients, 77% were non insulin dependent, 11% were non insulin dependent on insulin, and 12% were insulin dependent.

The mean bed occupancy by patients with diabetes was 9.1%. This is a significant percentage and represents the potential daily inpatient workload for the DSNs. In the authors’ hospital, this translated to an average of 34 diabetic inpatients at any one time.

The response rate to the DSN questionnaire was 100%. The responses indicated that 24 nurses (63%) spent between 1 and 5 hours per week on the wards. Three nurses (8.0%) spent more than 16 hours per week on the wards, but these were specifically employed for inpatient care.

The Wessex DSNs were asked to identify categories of patients whom they would wish to see during an inpatient episode. Their responses are shown in Table 2.

The majority of admissions in the inpatient audit were emergencies (83%) which were predominantly due to acute medical conditions. This finding is similar to that of Child et al (1991). In addition, five patients were newly diagnosed with diabetes, although this was not the primary reason for admission. Only two admissions were related to diabetes control: one had diabetic ketoacidosis secondary to a urinary tract infection and the other had ‘poor diabetic control’ secondary to an asthma attack.

These findings indicate that patients falling within the categories deemed as requiring DSN involvement are not necessarily being admitted. The decreasing number of admissions related to diabetes control could be seen as a reflection of the rise in number of DSN posts, as suggested by Harrower et al (1993) and Williams et al (1994).

With the drive towards teaching blood glucose monitoring (BGM) to facilitate the achievement of good control, it was surmised that patients who were monitoring their blood glucose before admission would

have a higher incidence of good control (Tables 3 and 4). The results show that 64% had good control before admission. Of these, 43% were performing BGM. However, the 3% who were performing urine testing and the 17% who were not monitoring at all still achieved good control.

There was no significant difference between the glycated haemoglobin (HbA_1c) obtained from laboratory records before admission and the HBA_1c performed on the day of audit (mean = 7.4% for both). The levels of diabetic control between those patients known to the diabetes centre (DSNs) and those who were not (Table 5) were compared.

There was no difference in the number of patients with HbA_1c in the good control range (HbA_1c 3.2-7.9%) (38 in each group). However, the number of patients in the poor control range (HbA_1c >8.0%) known to the diabetes centre was higher. This is assumed to be due to the increased referral of patients with poor control to the diabetes nursing service for intervention.

HbA_1c determined on the day of audit (Table 5) demonstrated that 58% of inpatients had good control (and therefore did not require DSN intervention).

Nineteen per cent of the Wessex DSNs said they would wish to see patients following a change in diabetes treatment, while a further 17% specified they would do so only if conversion to insulin had taken place during the inpatient stay. This may be appropriate as the inpatient audit showed that 61% of patients required no alteration to their preadmission treatment.

The inpatient audit undertaken in the authors’ Trust demonstrated that no patients were converted to insulin during their admission. However, during the same 4-month period, 16 outpatients were converted to insulin via the diabetes centre.

Thirty-one of the 138 patients (22%) were commenced on either subcutaneous or intravenous insulin at some stage during their admission and were subsequently returned to their preadmission medication. Only 8% of inpatients were newly commenced on oral hypoglycaemic agents and only 12% had minor alterations to their preadmission treatment.

This correlates well with the proportion of patients with good control (58%). There was no significant difference between the number of patients admitted who were known to the diabetes centre (71; 51%) and those who were not (67; 49%).

Sixty-three per cent of patients were referred to the DSN according to established referral criteria (Figure 2). The development of the referral proforma has increased the referral rate; however, the design allows the DSN to prioritise those patients who need to be seen. Intervention by the DSN would have taken place if appropriate.

Thirty-seven per cent of inpatients were not referred. This would appear appropriate as 58% had good control and 61% of patients had no changes in their treatment, indicating that intervention by the DSN may not have been necessary.

The majority of Wessex DSNs (83%) thought they could improve the service provided to the wards: 36% related it to an improved referral system, 33% to the need for more time but without specifying how they would utilise it, and 25% to the need for increasing staff education (multiple responses were received).

Conclusion
The main purpose of the audits was to identify potential inpatient workload and assess whether redistribution of DSN resources was required.

The time given to inpatient care by the authors is similar to that given by the majority (66%) of the Wessex DSNs. However, it is acknowledged that this may not be the case nationally. Similar audits would need to be performed in each region.

The authors have already spent much time improving the referral system in place at the Royal West Sussex Trust. This involved designing, in conjunction with ward staff, a referral form which was piloted in a number of areas. The design ensures that patient knowledge regarding diabetes control and self-management skills are assessed at ward level. The assessment areas covered are felt to be within the capability of all trained nurses. 

In the authors’ Trust, it is recommended that all inpatients with diabetes are referred to the diabetes nursing services. Patients are then seen by the DSN if necessary, either before or after discharge. Recommending the referral of all inpatients encourages nurses working in the acute unit to perform an assessment of the patient with diabetes. As the inpatient audit demonstrated, 67% of patients are being referred according to the authors’ policy.

The inpatient audit demonstrated that a large percentage of patients are well controlled and have no changes made to their treatment during an admission. The decreasing trend for patients admitted with diabetes-related problems, such as diabetic ketoacidosis and poor control, probably reflects the outpatient DSNs’ work. The percentage of time currently spent by DSNs on the wards would therefore appear to be appropriate.

To ensure that referrals are appropriate (and timely) it is recommended that DSNs:

• Establish a referral policy/protocol
• Design a referral system that encourages trained nurses to acknowledge their role in caring for diabetic patients in the acute unit.
• Evaluate the effectiveness of the above recommendations by audit.
• Provide ward-based resources so that if deficiencies in patient knowledge and self-caring skills are identified, these may be addressed by the ward nurse performing the assessment.