A commonly held view amongst experienced geriatricians is that strict adherence to onerous diabetic diets with maintenance of tight blood sugar ranges does not confer morbidity or mortality benefits to very old adults, and comes at a cost to patient quality of life. Clinical decisions to relax glycaemic control in the context of significant comorbidity occasionally leads to conflict with other clinicians or patients and their families. To date, there had been no large studies to help inform such management decisions in very old adults. The present study did not find that diabetes was associated with increased odds of mortality in people aged over 65 years with an unadjusted odds ratio of 0.89 (p<0.001). Diabetes diagnosis has an interaction with age. Accumulating age reduces its impact to the point that a diagnosis of diabetes may predict a better outcome in the very old adults (aged more than 80 years), see Figure 1a. This may be due to immortal time bias (only the healthiest diabetics survive to ages beyond 80 years) [17, 18]. At the very least, it confirms that diabetes in very old adults does not convey a worse prognosis.
The transition point to the very old adult is arbitrary. By using the median age of the cohort (80 years) as this study’s transition point, additional analysis, restricted to the oldest, showed no effect of diabetes diagnosis on mortality (see Table 3). Most studies that deal with mortality and diabetes are longitudinal studies [19-23]. There are only a few studies that explore in-hospital mortality associated with diabetes, and these are disease-specific [24-26]. Two of the studies showed no effect on inpatient mortality [24, 26]. The third showed an odds ratio of 1.31 (1.04-1.65) for mortality in people with diabetes with foot disease [25]. There may be disease-specific subsets that are at higher risk, but this was not explored in the present study.
Similar to mortality, diabetes diagnosis had less impact on the length of stay (LOS). Greater LOS may be used as an indicator of greater physical and/or psychological morbidity. Very old adults with diabetes had shorter lengths of stay (see Figure 1b) despite people with diabetes having overall longer LOS. The impact of diabetes on LOS is consistent with other studies [24, 26-28]. The present study differs from these works by including only people over 65 years and adjusting for the confounding of disease burden by using the CCI. Furthermore, also consistent with these studies is the magnitude of the effect on LOS, which is small.
Hospital-acquired adverse events were higher in the diabetic cohort. It was not modified when adjusting for age. This study used a validated but not extensively used method of detection of hospital-acquired diagnosis [16, 29-33]. The present study also found a much higher proportion of HADX overall, probably related to the examination of an older cohort. By contrast, Cromarty et al. were able to show that 29.3% of people with diabetes developed hospital-acquired events compared to 13% of non-diabetics [29]. .
This study demonstrated that the likelihood of re-admission within 28 days of discharge was associated with the diagnosis of diabetes, and these odds were not influenced with increasing age. Caughey et al. identified older people (aged over 85 years) with comorbidities as those most likely to be re-admitted within 30 days [34]. The present study adjusted for the presence of comorbidity and found little difference, providing some support for their result. Dungan identified those with poor glycaemic control as those most likely to be re-admitted[35]. Clinical measures were not undertaken in the present work.
The present study does not show why the diagnosis of diabetes does not appear to have an impact on mortality in very old adults. However, it does support clinical decisions to relax glycaemic control goals in this age group, as has been more broadly recommended in international guidelines [36]. There is evidence that rigid control may not have the benefits seen in younger people [37]. A converse viewpoint is that diabetes may not be as harmful in older people, and so its control does not need to be as tight. These considerations will enhance the adoption of a more measured approach to diabetic care in very old adults, particularly those with comorbidities, with less strict blood sugar ranges and a more liberal diabetic diet. These measures may in turn facilitate maintenance of independence and improved quality of life in this age group.
The current study has limitations. It is a retrospective audit of hospital administrative data. Hospital databases have not been designed for clinical investigations. However, the large amounts of data might be used for association. Several validation studies of hospital discharge data exploring diabetes diagnosis (as used in the present study) have been conducted with a diabetes diagnosis giving positive predictive values from 59% to 93% and negative predictive values from 91% to 99% [38]. These studies suggest that not having a label of diabetes is more accurate. Improved accuracy of the diagnosis of diabetes can occur by linking hospital datasets with other sets such as prescription databases [39]. Linkage studies were beyond the scope of this study but would merit further research. Moreover, a validation sub-study may be useful. Studies also warn about changes in coding rules occurring over time, such as the changes that occurred in the definitions of diabetes in 2011 [40]. This change has resulted in a decreased level of reporting. This study commenced in 2012 for that reason.
This study chose to analyse each admission as a separate event for its primary outcomes. Thus, individuals may be represented multiple times. However, including or excluding those with multiple admissions could be debated. This study analysed with and without multiple admissions, and with and without prior admission (see Table 3). There was no material difference across the different models.
This study did not use specific clinical measures such as medication usage, glycaemic control and measures of frailty. Several small works have examined hospital outcome based on these clinical measures [41-44]. Prospective studies that explore the effects of glycaemic control and frailty on hospital outcomes are needed.