Diabetes Diagnosis, Resource Utilization, and Health Outcomes

Objective: To determine the effect of a clinical diagnosis of diabetes mellitus (DM) on healthcare utilization and health outcomes.

Study Design: Cohort study.

Methods: A total of 197 United Kingdom family practices with 4974 subjects (mean age, 62.8 years; 52.2% men) with type 2 DM and 9948 matched nondiabetic control subjects. Healthcare utilization and the occurrence of complications were estimated from 2 years before to 2 years after the first clinical diagnosis of DM.

Results: From 24 months before the DM diagnosis, primary care consultations were increased in prediagnosis cases compared with controls (relative rate [RR], 1.31; 95% confidence interval [CI], 1.27-1.35), as were emergency and hospital care consultations, hospital specialist referrals, and prescription drug items. At diagnosis of DM, utilization of all forms of healthcare was increased (RR, 4.27; 95% CI, 4.17-4.36 for primary care consultations; RR, 2.49; 95% CI, 2.46-2.52 for prescription drug items). In the quarter following diagnosis, healthcare utilization was increased for acute myocardial infarction (RR, 6.29; 95% CI, 2.69-14.73), cerebrovascular disease (RR, 5.14; 95% CI, 3.37-7.84), ischemic heart disease (RR, 3.65; 95% CI, 2.77-4.80), and peripheral nerve disorders (RR, 5.01; 95% CI, 2.81-8.95). First diagnoses of myocardial infarction, cerebrovascular disease, and peripheral nerve disorders were increased during the period from 6 months before to 6 months after diagnosis.

Conclusions: Clinical diagnosis of DM is often the end of a process leading to established complications and is associated with greatly increased utilization of care. This adds to the justification of strategies for earlier detection of hyperglycemic states.

(Am J Manag Care. 2008;14:32-38)

This study aimed to evaluate the ways in which a clinical diagnosis of DM isassociated with healthcare utilization and the occurrence of clinical complications

Family practices were selected if they continuously provided data from January 1, 2001, to December 31, 2005. There were 1 283 429 patients 100 years or younger who were continuously registered with 197 family practices during this time. From this cohort, medical diagnostic codes for DM and prescriptions for oral hypoglycemic drugs or insulin were used to identify cases with DM. The DM diagnosis date was defined as the earlier of the first DM medical code or the first DM prescription date. Cases with DM were selected if their diagnosis date was between January 1, 2003, and December 31, 2003 (5635 cases), and if they were aged between 30 and 89 years at diagnosis (5091 cases). Two controls were selected for each case (matching for age, sex, and family practice) from among subjects who had never been diagnosed as having DM or prescribed oral hypoglycemic drugs or insulin. Twenty-two cases were excluded because there were insufficient matched controls. Finally, we included only 4974 patients who had never been diagnosed as having type 1 DM.

Data Analysis

Data were analyzed for 4974 subjects (mean age, 62.8 years [age range, 30-89 years]; 52.2% men), drawn from 197 family practices, who were diagnosed as having type 2 DM during 2003. There were 9948 control subjects, never diagnosed as having DM, who were matched for age, sex, and family practice.

Figure 1 shows the utilization of healthcare before and after the diagnosis of DM compared with that of nondiabetic control subjects. In the quarter from 24 to 22 months before the diagnosis of DM, the RRs were 1.31 (95% CI, 1.27-1.35) for primary care consultations, 1.23 (95% CI, 1.08-1.39) for emergency and hospital care, 1.24 (95% CI, 1.14-1.35) for hospital specialist referrals, and 1.55 (95% CI, 1.52-1.57) for prescription drug items. The relative excess of medical care utilization continued through the prediagnosis period and then showed a marked increase, reaching a maximum in the quarter following diagnosis of DM. At this time, the RR for primary care consultations was 4.27 (95% CI, 4.17-4.36) times higher than that for controls; the RRs were 2.41 (95% CI, 2.18-2.66) times higher for emergency and hospital care, 2.97 (95% CI, 2.79-3.16) times higher for hospital specialist referrals, and 2.49 (95% CI, 2.46-2.52) times higher for prescription drug items. By 22 to 24 months after diagnosis, utilization of medical care had declined but remained higher than that in controls and higher than that in the prediagnosis period. Compared with controls, the RRs for cases were 1.64 (95% CI, 1.60-1.68) for primary care consultations, 1.42 (95% CI, 1.26-1.59) for emergency and hospital care, 1.50 (95% CI, 1.42-1.57) for hospital specialist referrals, and 2.10 (95% CI, 2.08-2.13) for prescription drug items. Table 1 gives the absolute rates and RRs of healthcare utilization for 3 broad periods before, during, and after the diagnosis of DM confirming these findings.

Figure 2 shows the RRs of healthcare utilization events, combined across various types of care, for 4 different conditions. These analyses of healthcare utilization for specific groups of conditions showed that utilization of medical care was increased from 24 to 22 months before DM diagnosis for ischemic heart disease (RR, 2.13; 95% CI, 1.60-2.84) but not for acute myocardial infarction (RR, 1.33; 95% CI, 0.47- 3.75), cerebrovascular disease (RR, 1.00; 95% CI, 0.57-1.76), or peripheral nerve disorders (RR, 1.33; 95% CI, 0.68-2.62). There were statistically significant peaks in the utilization of care for these conditions at the time of diagnosis. In the quarter following diagnosis of DM, the RRs were 3.65 (95% CI, 2.77-4.80) for ischemic heart disease events, 6.29 (95% CI, 2.69-14.73) for acute myocardial infarction, 5.14 (95% CI, 3.37-7.84) for cerebrovascular disease, and 5.01 (95% CI, 2.81-8.95) for peripheral nerve disorders. However, utilization of care for these conditions declined to levels similar to those of controls by 22 to 24 months after diagnosis (RR, 1.05; 95% CI, 0.70-1.58 for ischemic heart disease; RR, 2.01; 95% CI, 0.41-9.96 for acute myocardial infarction; RR, 1.47; 95% CI, 0.81-2.65 for peripheral nerve disorders; and RR, 1.64; 95% CI, 0.93-2.87 for cerebrovascular disease). Table 2 summarizes how, in the period from 6 months before to 6 months after diagnosis, the absolute rates and RRs of new diagnoses of each condition were statistically significantly elevated. New ischemic heart disease and peripheral nerve disorder events were increased in the postdiagnosis period, but cerebrovascular disease events and acute myocardial infarctions were not.

DISCUSSION

We hypothesized that the clinical recognition of DM may be an event of limited significance in the course of a condition that is already present. Our results show that this is not the case. A clinical diagnosis of DM is associated with a profound increase in healthcare utilization and with increased occurrence of newly diagnosed ischemic heart disease, acute myocardial infarction, cerebrovascular disease, and peripheral nerve disorders. A clinical diagnosis of DM is a costly occurrence in terms of healthcare resource utilization and associated adverse health outcomes of DM. However, the nature of the interaction between health services and affected individuals is complex. The occurrence of illness may lead to contacts with health services at which DM may be diagnosed. Conversely, a diagnosis of DM may lead to the detection of comorbid conditions that were already present. In either case, there will be delivery of needed care for treatment of present conditions and prevention of future complications. Clinical practice recommendations for good-quality DM care require that utilization of care is increased to facilitate assessment for complications and intervention to control hyperglycemia, hypertension, and lipid disorders. Therefore, even when diabetic complications are not present, some increase in utilization of care around the time of DM diagnosis may be anticipated. Furthermore, there may be subgroups of patients in whom the occurrence of complications, or the increase in utilization of care associated with DM diagnosis, is more or less evident.

Comparison With Other Studies

This was a population-based study among a large number of family practices drawn from across the United Kingdom. The large sample size yielded precise estimates. Cases and controls were matched for age, sex, and family practice, but unmeasured variables such as cigarette smoking, obesity, and individual-level socioeconomic status may have accounted for observed differences. Smoking,^19,20 obesity,^21,22 and lower socioeconomic status²³ are associated with an increased frequency of DM and, in some studies,^24,25 with greater utilization of primary care services. However, these characteristics may be causally associated with prediabetes and DM and are not true confounders.

Implications for Chronic Illness Care

We thank the staff of EPIC-UK for facilitating access to The Health Improvement Network database.

2. Gillies CL, Abrams KR, Lambert PC, et al. Pharmacological and lifestyle interventions to prevent or delay type 2 diabetes in people with impaired glucose tolerance: systematic review and meta-analysis. BMJ. 2007;334:e299.

4. Gulliford MC, Charlton J, Latinovic R. Increased utilization of primary care five years before diagnosis of type 2 diabetes: a matched cohort study. Diabetes Care. 2005;28:47-52.

6. Icks A, Haastert B, Giani G, Rathmann W. Incremental prescription and drug costs during the years preceding diabetes diagnosis in primary care practices in Germany. Exp Clin Endocrinol Diabetes. 2006;114:348-355.

8. Tata LJ, Fortun PJ, Hubbard RB, et al. Does concurrent prescription of selective serotonin reuptake inhibitors and non-steroidal antiinflammatory drugs substantially increase the risk of upper gastrointestinal bleeding? Aliment Pharmacol Ther. 2005;22:175-181.

10. StataCorp LP. STATA Statistical Software, Release 9. College Station, TX: StataCorp LP; 2006.

12. Kohner EM, Aldington SJ, Stratton IM, et al. United Kingdom Prospective Diabetes Study, 30: diabetic retinopathy at diagnosis of non–insulin-dependent diabetes mellitus and associated risk factors. Arch Ophthalmol. 1998;116:297-303.

14. Centers for Disease Control and Prevention (CDC). Prevalence of diabetes and impaired fasting glucose in adults: United States, 1999-2000. MMWR Morb Mortal Wkly Rep. 2003;52:833-837.

16. Rijkelijkhuizen JM, Nijpels G, Heine RJ, Bouter LM, Stehouwer CDA, Dekker JM. High risk of cardiovascular mortality in individuals with impaired fasting glucose is explained by conversion to diabetes: the Hoorn Study. Diabetes Care. 2007;30:332-336.

18. Gulliford MC, Latinovic R, Charlton J, Hughes RAC. Increased incidence of carpal tunnel syndrome up to 10 years before diabetes diagnosis. Diabetes Care. 2006;29:1929-1930.

20. Rimm EB, Manson JE, Stampfer MJ, et al. Cigarette smoking and the risk of diabetes in women. Am J Public Health. 1993;83:211-214.

22. McLaren L. Socioeconomic status and obesity. Epidemiol Rev. 2007;29:29-48.

24. van Doorslaer E, Wagstaff A, van der Burg H, et al. Equity in the delivery of health care in Europe and the US. J Health Econ. 2000;19:553-583.

26. Waugh N, Scotland G, McNamee P, et al. Screening for type 2 diabetes: literature review and economic modelling. Health Technol Assess. 2007;11:1-144.

Download PDF: Diabetes Diagnosis, Resource Utilization, and Health Outcomes