- Email: [email protected]
Quality of Diabetes Care in the Canadian Forces
Can J Diabetes 38 (2014) 11e16
Contents lists available at ScienceDirect
Canadian Journal of Diabetes journal homepage: www.canadianjournalofdiabetes.com
Original Research
Quality of Diabetes Care in the Canadian Forces Amole Khadilkar MD a, *, Jeff Whitehead MD a, c, Monica Taljaard PhD a, b, Doug Manuel MD a, b a b c
Department of Epidemiology and Community Medicine, University of Ottawa, Ottawa, Ontario, Canada Clinical Epidemiology Program, Ottawa Hospital Research Institute, Ottawa, Ontario, Canada Department of National Defence, Canadian Forces Health Services, Ottawa, Ontario, Canada
a r t i c l e i n f o
a b s t r a c t
Article history: Received 1 February 2013 Received in revised form 21 July 2013 Accepted 18 August 2013
Background: Published data on quality of care indicators from various countries indicate the challenges of providing high-quality diabetes care. The objective of this study was to evaluate the quality of care provided to members of the Canadian Forces (CF) who have diabetes, by determining the extent to which healthcare providers adhere to recommendations outlined in the 2008 Canadian Diabetes Association (CDA) clinical practice guidelines. Methods: All 14 CF bases meeting eligibility criteria were included in the evaluation. Cases of diabetes were ascertained based on laboratory criteria. Adherence to 21 CDA guideline recommendations was evaluated following a review of patient medical records. Results: The CF demonstrated high adherence (>75%) with 9 recommendations, moderate adherence (50% to 75%) with 7 recommendations and low adherence (<50%) with 5 recommendations. Most notably, there were 4 recommendations for which adherence was greater than 90%. The mean rate of adherence with all applicable recommendations per patient was 60.3% (95% Confidence Interval [CI], 59.0% to 61.6%). CF adherence rates were generally similar to or better than comparable rates in the civilian population within Canada and other industrialized countries. Conclusions: It is unclear what accounts for the favourable quality of diabetes care in the CF Health Services, but this highly structured practice setting has a number of features that distinguish it from provincial healthcare systems. Several strategies can be considered to improve diabetes care even further, including providing feedback to physicians about their performance, promoting the use of diabetes care flow sheets and creating a diabetes registry. Ó 2014 Canadian Diabetes Association
Keywords: quality of care Clinical Practice Guidelines Canadian Forces
r é s u m é Mots clés : qualité des soins lignes directrices de pratique clinique Forces canadiennes
Introduction : Les données publiées sur les indicateurs de la qualité des soins de plusieurs pays montrent les défis liés à la prestation de soins de grande qualité aux diabétiques. L’objectif de cette étude était d’évaluer la qualité des soins fournis aux membres des Forces canadiennes (FC) qui ont le diabète en déterminant dans quelle mesure les prestataires de soins respectent les recommandations définies dans les lignes directrices de pratique clinique 2008 de l’Association canadienne du diabète (ACD). Méthodes : Les 14 bases des FC répondant aux critères d’admissibilité ont été incluses dans l’évaluation. Les cas de diabète ont été établis selon les critères de laboratoire. L’observance des 21 recommandations provenant des lignes directrices de l’ACD a été évaluée d’après une revue de dossiers médicaux. Résultats : Les FC ont démontré une observance élevée (> 75 %) si 9 recommandations avaient été suivies, une observance modérée (50 % à 75 %) si 7 recommandations avaient été suivies et une faible observance (< 50 %) si 5 recommandations avaient été suivies. Plus particulièrement, il y a eu 4 recommandations pour lesquelles l’observance a été plus grande que 90 %. Le taux moyen d’observance de toutes les recommandations applicables par patient a été de 60,3 % (intervalle de confiance [IC] à 95 %, 59,0 % à 61,6 %). Les taux d’observance des FC ont généralement été similaires ou meilleurs que les taux comparables de la population civile de tout le Canada et d’autres pays industrialisés. Conclusions : Il est difficile de savoir ce qui relève d’une qualité de soins favorable dans les Services de santé des FC, mais ce cadre de pratique hautement structuré a de nombreuses caractéristiques qui le distinguent des systèmes de soins de santé provinciaux. Plusieurs stratégies peuvent être considérées
* Address for correspondence: Amole Khadilkar, MD, Department of Epidemiology and Community Medicine, Faculty of Medicine, University of Ottawa, 451 Smyth Road, Room 3105B, Ottawa, Ontario K1H 8M5, Canada. E-mail address: [email protected] 1499-2671/$ e see front matter Ó 2014 Canadian Diabetes Association http://dx.doi.org/10.1016/j.jcjd.2013.08.264
12
A. Khadilkar et al. / Can J Diabetes 38 (2014) 11e16
pour améliorer encore davantage les soins du diabète, y compris l’offre d’une rétroaction aux médecins sur leur performance, la promotion de l’utilisation des organigrammes des soins diabétologiques et la création d’un registre du diabète. Ó 2014 Canadian Diabetes Association
Introduction Between 2008 and 2009, there were an estimated 2.4 million Canadians with diabetes, representing 6.8% of the population (1). Diabetes is the 6th leading cause of death in Canada, with cardiovascular disease representing a major cause of morbidity and mortality among affected individuals (2,3). There is a 2- to 4-fold increased risk of cardiovascular events among people with diabetes compared to those without diabetes, an increase that persists even after accounting for other traditional risk factors (4). Additional complications include blindness, kidney disease, foot ulcers and amputations. The annual cost of diabetes to the Canadian healthcare system is projected to exceed $19 billion by 2020 (5). Indicators of the quality of diabetes care internationally have demonstrated that there are significant challenges to providing high-quality diabetes care (6e10). With the prevalence of diabetes increasing globally due to an aging population and increasing levels of obesity and physical inactivity, the quality of diabetes care has never been a more important issue (11). The objective of this study was to evaluate the quality of care provided to members of the Canadian Forces (CF) who have diabetes, by determining the extent to which health care providers adhere to recommendations outlined in the 2008 Canadian Diabetes Association (CDA) clinical practice guidelines (12).The recommendations proposed in the CDA guidelines were based on the best available evidence as well as expert opinion on best practice. They shared much in common with the American Diabetes Association (ADA) clinical practice guidelines and the UK’s National Institute for Health and Clinical Excellence (NICE) clinical practice guidelines; although there were notable differences as well (13,14). A secondary objective was to compare the quality of diabetes care in the CF with national and international estimates for the general population.
notes or periodic health assessments (diagnostic assessment extract) or who were identified from a pharmacy database as having been prescribed diabetes medication and/or glucose testing strips in the past year. The diagnostic assessment extract contained the diabetes-specific ICD-10 codes (E10-E14) listed for every relevant clinic visit as early as July 2010 up to December 19, 2011. From this initial list, eligible patients for this study were identified by reviewing laboratory records. Patients were included in this study if they had plasma glucose levels in the diabetes range on 2 separate occasions as per clinical practice guidelines (fasting plasma glucose 7.0 mmol/L, random plasma glucose 11.1 mmol/L or 2-hour plasma glucose 11.1 mmol/L after a 75 g oral glucose tolerance test) (12). As the 2008 CDA clinical practice guidelines did not include HbA1c thresholds in their diagnostic criteria, HbA1c levels were not considered. Note that while patients may be prescribed medications, like Metformin for conditions such as pre-diabetes and polycystic ovarian syndrome or require glucose testing strips for monitoring gestational diabetes, the requirement for laboratory evidence of diabetes meant that these conditions were excluded from consideration. Because the CF does not accept applicants with pre-existing diabetes, CF members currently with diabetes must have been diagnosed while in service and would, therefore, have laboratory results corresponding to that diagnosis in their medical records. Quality of care indicators The criteria used to assess quality of care were based on adherence to CDA guideline recommendations in 2 areas: 15 processes of care and 6 intermediate outcomes. Guideline recommendations that should be “considered” were not included in the evaluation as they are open to clinical judgement.
Methods
Process of care measures
A retrospective medical case review of all eligible CF members with diabetes was conducted over a 3-year period to examine adherence to 21 CDA guideline recommendations.
In terms of processes of care, the 2008 CDA guidelines recommended that patients with diabetes should have the following: a hemoglobin A1c test at least every 6 months; an annual test for urine protein excretion (random urine albumin-creatinine ratio [ACR]); an annual kidney function test (serum creatinine converted to an estimated glomerular filtration rate [eGFR]); a blood pressure check at every visit; an annual foot exam with additional testing for peripheral neuropathy; a dilated eye exam every 1 to 2 years; a fasting lipid profile every 1 to 3 years and a baseline EKG in those over 40 years of age (12). An EKG should be repeated every 2 years in those who are at high risk for cardiovascular events (7). Patients at high risk for cardiovascular events were defined as males 45 years old and as females 50 years old (12). All other patients with diabetes were defined as moderate risk. Additional recommendations included an annual influenza vaccination, encouragement to stop smoking, the use of metformin as the initial oral anti-hyperglycemic agent, and the prescription of an angiotensin-converting enzyme (ACE) inhibitor or an angiotensin receptor blocker (ARB) for those with persistent microalbuminuria (ACR 2.0 mg/mmoL in males or 2.8 mg/mmoL in females), even in the absence of hypertension (12). According to the 2008 CDA clinical practice guidelines, an ACE inhibitor or ARB should also be prescribed for those at high risk of cardiovascular events (12).
Study population The study population consisted of members of the CF Regular Force aged 18 to 60 years with diabetes who were located at the 14 bases that incorporated progress notes into the electronic medical record (EMR) for 6 months or more as of December 19th, 2011. There was a total of 32 603 members of the CF Regular Force at these bases, which represents roughly half of the overall CF Regular Force population. The bases included members of the army, navy and air force and were located in all regions of Canada except the Maritime provinces, which did not meet the electronic medical record (EMR) criteria for inclusion, and the territories, where there are no CF bases. Case ascertainment of diabetes CF members with diabetes were identified by extracting a list of patients at the 14 selected bases who either had a diagnosis of diabetes documented in the EMR within the physicians’ progress
A. Khadilkar et al. / Can J Diabetes 38 (2014) 11e16
Intermediate outcomes In terms of intermediate outcomes, the 2008 CDA clinical practice guidelines indicated that among patients with diabetes, hemoglobin A1c levels should be 7.0%, blood pressure should be <130/80 mm Hg and, in those at high risk for cardiovascular events, low-density lipoprotein (LDL) should be 2.0 mmol/L with a secondary target being a total cholesterol:high density lipoprotein (TC:HDL) ratio of less than 4.0 (12). In patients with an ontreatment LDL level of 2.0 to 2.5 mmol/L, clinical judgement should be used to decide if additional therapy is necessary (12). Given that additional therapy for LDL levels between 2.0 mmol/L and 2.5 mmol/L was at the discretion of physicians, they were given the benefit of the doubt and assessed as adhering to guideline recommendations if LDL levels were 2.5 mmol/L. In those patients who were at moderate risk for cardiovascular events, the Canadian dyslipidemia guidelines at the time indicated that cholesterol lowering treatment should be initiated if LDL was >3.5 mmol/L and the TC:HDL ratio was 5.0 (33). Data extraction One abstractor (AK) extracted the data from the medical records. A second abstractor (JW) independently reviewed a random sample of 10% of the charts to determine the inter-rater reliability for 10 of the most important guideline recommendations using the kappa statistic.
Statistical analysis Adherence to each of the 21 guideline recommendations was calculated as proportions with 95% large-sample confidence intervals (CIs). In addition, the mean rate of adherence with the entire panel of applicable recommendations per patient was calculated with 95% CIs. Indirect age and sex standardization of diabetes prevalence rates in the CF was conducted using age and sex-specific Canadian prevalence rates from 2008 to 2009 (34). All statistical analyses were performed using Stata software (version 11.0; StataCorp LP, College Station, TX, USA).
National and international comparisons Results were compared with quality of care data from Alberta’s primary care networks, which were obtained from a large cohort study involving 77464 diabetes patients (15). Primary care networks (PCN) represent a multidisciplinary practice model with a unique funding scheme in which health professionals are paid on a traditional fee-for-service basis and an additional $50 per patient is provided to the PCN to support its ongoing activities (15). Aside from the Canadian Community Health Survey (CCHS), the PCN cohort study is the largest Canadian study reporting on quality of diabetes care. The CCHS had a much more narrow focus with fewer indicators (9). In addition to the PCN quality of care data, comparisons were made with international estimates obtained from the published literature and in the case of the CCHS, the unpublished literature (8e10,15).
Ethics clearance The study was approved by the Director of Access to Information and Privacy (DAIP), as well as Veritas, an independent review ethics board accredited by the Association for the Accreditation of Human Research Protection Programs (AAHRPP).
13
Table 1 Characteristics of the study population Frequency (%)* Gender Males Females Age Mean (SD) Age <40 years Age 40e49 years Age 50e60 years Rank Officers (university education, higher rank, higher income) Noncommissioned members (less education, lower rank, lower income) Smoking status Smokers Nonsmokers Type of diabetes Diet-controlled type 2 diabetes Type 2 diabetes on oral agents Type 2 diabetes on insulin Type 1 diabetes
383 (95.8%) 17 (4.3%) 46.2 53 218 129
(6.3) (13.3%) (54.5%) (32.3%)
45 (11.3%) 355 (88.8%)
100 (25.4%) 294 (74.6%) 32 326 33 9
(8.0%) (81.5%) (8.3%) (2.3%)
SD, standard deviation. * Unless otherwise indicated.
Results The characteristics of the study population are presented in Table 1. A total of 400 diabetes patients were identified using the diagnostic assessment extract and the pharmacy database at the 14 bases included in the study, giving a total diabetes prevalence of 1.2% (95% CI, 1.1e1.3%). One factor contributing to the relatively low diabetes prevalence is the younger population in the CF, two-thirds of who are under the age of 40 years. It is well recognized that the risk of diabetes increases with age. Using the age and sex-specific diabetes prevalence rates in the Canadian population as a reference standard, the total expected number of cases among females in the CF is 106, which is more than the observed number of 17 by a factor of 6 (indirectly standardized prevalence ratio 0.16, 95% CI 0.094e0.26). The total expected number of cases among males is 730, which is more than the observed number of 383 by a factor of almost 2 (indirectly standardized prevalence ratio 0.52, 95% CI 0.47e0.58). Therefore, even after age standardization, diabetes prevalence rates in the CF appear to be lower than that of the general Canadian population. Increased physical activity among members of the CF may be a possible explanation for some of the decreased risk relative to the general population. In addition, the presence of comorbidities among members of the CF with diabetes could be grounds for early medical release, thus reducing the number of people captured in our prevalence statistics. Finally, it is known that women in the CF have significantly lower rates of obesity (17%) compared to their male counterparts (25%), which may account for their particularly lower risk. The level of inter-rater agreement for individual quality of care indicators was high, with the kappa statistic ranging from 0.75 to 0.96. The adherence rates for the 21 CDA guideline recommendations considered in this study are shown in Table 2. For the CF regular force, there were 9 recommendations for which adherence was greater than 75%, including 4 recommendations for which adherence was greater than 90%. These 4 recommendations included using metformin as the initial oral anti-hyperglycemic agent, lipid profile testing every 1 to 3 years, performing a baseline EKG in those over 40 years of age, and prescribing an angiotensinconverting enzyme (ACE) inhibitor or angiotensin receptor blocker (ARB) in patients with persistent microalbuminuria. There
14
A. Khadilkar et al. / Can J Diabetes 38 (2014) 11e16
Table 2 Adherence with the Canadian Diabetes Association clinical practice guideline recommendations among patients within the CF Regular Force and the civilian population (point estimate with 95% confidence interval where available) Recommendation Processes of Care Metformin should be the initial oral anti-hyperglycemic agent prescribed Lipid profile every 1 to 3 years Baseline EKG in patients over 40 years of age ACE inhibitor or ARB prescription in patients with persistent microalbuminuria Smoking cessation advice Eye exam every 1e2 years Annual eGFR HbA1c every 6 months ACE inhibitor or ARB prescription in patients at high risk Annual ACR Follow up EKG every 2 years in patients at high risk Annual influenza vaccination Annual foot exam Annual peripheral neuropathy testing. Blood pressure check at every visit Intermediate outcomes LDL 3.5 mmol/L (moderate risk) TC:HDL < 5.0 (moderate risk) LDL 2.5 mmol/L (high risk) TC:HDL < 4.0 (high risk) HbA1c 7.0% Blood pressure <130/80
CF Regular Force (n¼400)
Alberta’s Primary Care Networks (n¼77464) [15]
Best of International estimates [8e10]
98.9% (97.8e100.0%)
84.3% (83.1e85.5%)
-
96.8% (95.0e98.5%)
63.5% (63.1e63.8%)*
71% to 85% (U.S.)*
96.4% (94.4e98.4%)
-
-
90.9% (82.1e99.8%)
79.3% (78.2e80.3%)
-
82.2% 77.1% 75.6% 75.3% 71.3%
32.7% (32.4e33.0%) * 70.0% (69.7e70.3%)* -
32% to 77% (Australia)y 75% to 85% (U.S.) 65% to 93% (Australia)* -
67.4% (62.8e72.0%) 67.1% (60.9e73.4%)
44.7% (44.4e45.1%) -
-
43.4% (38.5e48.2%) 15.9% (12.3e19.5% 10.3% (7.3e13.3%)
-
51% (Canada) -
7.3% (4.7e9.8%)
-
-
53.2% (52.8e53.6%) -
31% to 47% (U.S.)z 38% to 57% (Australia) 30.9% (Taiwan)
85.3% 74.8% 71.9% 57.5% 53.9% 37.7%
(74.6e89.8%) (73.0e81.3%) (71.4e79.9%) (71.0e79.5%) (65.8e76.9%)
(79.3e91.3%) (67.6e82.0%) (66.4e77.5%) (51.5e63.6%) (48.8e59.0%) (32.7e42.8%)
ACE, angiotensin converting enzyme; ARB, angiotensin receptor blocker; ACR, albumin-creatinine ratio; EKG, electrocardiogram; LDL, low-density lipoprotein; TC:HDL, total cholesterol to high-density lipoprotein ratio; HbA1c, hemoglobin A1c. * Annual estimate. y Biennial estimate. z LDL<2.6 mmol/L in all patients regardless of risk.
was moderate adherence for 7 recommendations with rates ranging between 50% and 75%. There was low adherence (below 50%) for 5 recommendations. The mean adherence with all applicable CDA recommendations per patient was 60.3% (95% CI, 59.0e61.6%). Based on the 7 recommendations for which comparable data were available, the CF generally showed better adherence rates than Alberta’s PCNs. Differences were positive for 6 recommendations ranging from 5.3% to 44.4%; for one recommendation (maintaining a HbA1c at 7.0%), adherence was similar in the 2 populations. Compared to the best of international estimates, the CF showed lower adherence for HbA1c testing, annual kidney function testing and the annual foot exam. However, this study considered a stricter recommendation of HbA1c testing in the last 6 months, whereas Australia and other countries evaluated HbA1c testing in the last year (10). Therefore, corresponding adherence rates would be expected to be higher in those countries. Furthermore, the Australian data that demonstrated higher adherence involved patients visiting specialist diabetes clinics, not primary care practices as was the case in the current study. A higher rate of annual foot exams was found in the Canadian Community Health Survey, which excluded members of the CF, but this represents patient self-reported data that is subject to information bias (9). Furthermore, we only included foot exams documented in the patient’s medical records, which may under-represent performance of the intervention. The guideline adherence rate for the remaining recommendations was the same or better in the CF
compared to countries with the best adherence rates based on international comparisons. Note the CF differs substantially from the populations being compared to in terms of age, gender, socioeconomic status and the presence of comorbidities. However, comparing the quality of care in these different populations based on adherence to the best practices outlined by the CDA has relevance. The CDA guideline recommendations apply equally to people with diabetes of all ages and genders, regardless of the presence of comorbidities and whether the patient is diet controlled or on insulin. Most of the indicators evaluated in this study were process of care measures, which physicians should observe regardless of the health status of the patient. It is acknowledged, however, that older individuals with longer durations of diabetes and multiple comorbidities may be more difficult to manage effectively. Adherence to target intermediate outcomes may be more difficult to achieve in such patients. Therefore, for these specific outcome measures, the comparisons between different populations should be interpreted with caution. Discussion Of the 21 CDA guideline recommendations considered, there was very high (>90%) or high adherence (>75%) for nearly half of the CDA recommendations considered; low adherence (<50%) was observed for only 5 recommendations. Based on comparison of available indicators with Alberta’s primary care networks, the quality of diabetes care in the CF appears to be better than that of
A. Khadilkar et al. / Can J Diabetes 38 (2014) 11e16
15
the civilian population in Canada and better than some international estimates. However, the mean adherence with all applicable CDA recommendations per patient was 60.3%, which is less than optimal. We were rigorous in our application of the CDA recommendations (e.g. adequate HbA1c status required HbA1c testing in the last 6 months). Had we followed a more flexible approach as in many other studies of diabetes care, our indicators would likely have shown an even more favourable picture of diabetes care in the CF. It is not clear what factors related to the Canadian Forces Health Services lend themselves to better diabetes practice. However, physicians caring for members of the CF are salaried, generally see fewer patients per day than physicians working in provincial health care systems, and have more time and resources to provide better comprehensive diabetes care. Medications and medical devices are provided free to CF members, removing financial barriers to treatment; Alberta residents must pay for medications on their own, typically through private insurance. The occupational repercussions of poorly controlled diabetes may motivate CF members to adhere more closely to a physician’s medical advice and engage in appropriate self-management. Finally, the use of the EMR likely plays a role in facilitating better diabetes practice in the CF. On the other hand, there are factors that make diabetes care in the CF more challenging, such as the periodic transfers of CF members to new bases and care teams and the relatively low volume of patients with diabetes (16,17). Despite the fact that the Canadian Forces Health Services demonstrated very good performance in many areas, there is room for improvement. Specific areas where there was low adherence (<50%) included receiving an annual influenza vaccination, maintaining a blood pressure at less than 130/80, getting an annual foot exam, having annual peripheral neuropathy testing and getting a blood pressure check at every medical visit. There are at least 5 quality improvement initiatives that are currently not used in the CF but may be considered to improve the suboptimal areas of diabetes care. First, performing an audit and providing feedback to health care providers about their performance is one useful strategy. Feedback has been shown to improve the quality of patient care, particularly in areas where performance is poor (18). Importantly, the feedback given to physicians should provide encouragement about the areas where they excel, of which several examples are identified in this study. Second, diabetes care flow sheets have been shown to be beneficial in improving guideline adherence, but are used by only a minority of CF physicians (19e21). More widespread use of diabetes care flow sheets would improve the documentation of important processes of care and intermediate outcomes and serve as a convenient reminder system for physicians. Third, clinical decision support tools that provide electronic reminders at the point of care have been shown to improve the quality of care, but they are not supported by the CF’s current EMR software (22e25). Fourth, making 10 g monofilaments available to all physicians for peripheral neuropathy testing is another potential strategy to improve care. Indeed, some CF physicians documented that 10 g monofilaments were not available as a reason for not performing peripheral neuropathy testing. Fifth, creating a diabetes registry may be an important step to further improving the quality of diabetes care in the CF (26,27). This appears feasible based on a case validation performed in conjunction with this study (not reported) and also supports a number of quality of care initiatives including several of the ones noted above.
likelihood of this is low. Second, it could be reasonably argued that the quality of care at the 14 bases selected for this study is not representative of the entire CF. However, these bases constituted almost 50% of the CF Regular Force. Moreover, it is unlikely that bases that had incorporated progress notes into the EMR for 6 months or more differed substantially from bases that did not, as the EMR was sequentially implemented on bases according to factors that were unlikely to be associated with quality of care such as geographic location and convenience. Third, for categorizing adherence to the annual influenza vaccination recommendation, we considered both the current season and the previous season since the current season was not over by the time of the chart review: December 19, 2011. While reasonable, this may have inflated the calculated vaccination rate, which was still well below the 80% target established for Canadians with chronic medical conditions (28). Fourth, in spite of the CDA recommendation to maintain a blood pressure <130/80, there is some controversy about how to define appropriate blood pressures in people with diabetes in light of several recent studies suggesting that intensive blood pressure control is not beneficial (29e32). Therefore, this particular CDA recommendation and the recommendation to have a blood pressure check at every visit may change in future updates of the guidelines. This has implications for the number of suboptimal areas of diabetes care in the CF. Adherence to target intermediate outcomes may be more difficult to achieve in populations with longer histories of diabetes and a higher rate of comorbidities. Therefore, for these specific outcomes, the comparisons between the CF and other populations should be interpreted with caution. Finally, quality of care may be affected by patient age, sex and socioeconomic status. Differences in these socio-demographic variables between the CF and the general population may, therefore, bias comparisons of quality of care between the 2 populations.
Limitations
Dr. Amole Khadilkar contributed substantially to the acquisition, analysis and interpretation of data and drafted the article. All authors contributed substantially to the conception and design of the research, revised the article critically for important intellectual content and gave final approval of the version to be published.
There were several limitations to our study. First, it is possible that we may have missed patients with diet-controlled diabetes who did not visit their physicians often. However, we believe the
Conclusion and future directions In conclusion, the CF fared reasonably well in spite of this rigorous quality of care assessment. Nearly half of the CDA recommendations evaluated had adherence rates over 75%. Moreover, the quality of diabetes care at the 14 bases investigated within the CF was generally similar to or better than the quality of care in the civilian population within Canada and other industrialized countries. Several quality improvement initiatives that may further improve diabetes care in the CF were identified with the creation of a diabetes registry being an important next step. Acknowledgements Special thanks to Major (Dr.) Stephen Cooper for guidance on the use of the EMR, to Robert Hawes for compiling the diagnostic assessment extract, to Janice Ma for providing access to the pharmacy database, to Zelma Buckley for delivering the paper charts and to Laura Bogaert for assistance with Stata programming. Author’s Disclosures There were no conflicts of interest with any of the authors. There were no financial disclosures. Author’s Contributions
16
A. Khadilkar et al. / Can J Diabetes 38 (2014) 11e16
References 1. Public Health Agency of Canada. Diabetes in Canada: Facts and figures from a public health perspective 2011. http://www.phac-aspc.gc.ca/cd-mc/publications/ diabetes-diabete/facts-figures-faits-chiffres-2011/chap1-eng.php. Accessed May 1, 2012. 2. Statistics Canada. Leading Causes of deaths in Canada 2008. http://www.statcan. gc.ca/pub/84-215-x/2011001/table-tableau/tbl001-eng.htm. Accessed May 1, 2012. 3. Miettinen H, Lehto S, Salomaa V, et al. Impact of diabetes on mortality after the first myocardial infarction. The FINMONICA Myocardial Infarction Register Study Group. Diabetes Care 1998;21:69e75. 4. Haffner SM, Lehto S, Rönnemaa T, et al. Mortality from coronary heart disease in subjects with type 2 diabetes and in nondiabetic subjects with and without prior myocardial infarction. N Engl J Med 1998;339:229e34. 5. Canadian Diabetes Association. The prevalence and costs of Diabetes 2008. http://www.diabetes.ca/files/prevalence-and-costs.pdf. Accessed May 1, 2012. 6. McGlynn EA, Asch SM, Adams J, et al. The quality of health care delivered to adults in the United States. N Engl J Med 2003;348:2635e45. 7. Harris SB, Ekoé JM, Zdanowicz Y, Webster-Bogaert S. Glycemic control and morbidity in the Canadian primary care setting (results of the Diabetes in Canada Evaluation study). Diabetes Res Clin Pract 2005;70:90e7. 8. Si D, Bailie R, Wang Z, et al. Comparison of diabetes management in five countries for general and indigenous populations: an internet-based review. BMC Health Services Research 2010;10:169. 9. Canadian Institute for Health Information (CIHI). Diabetes care gaps and disparities 2009. http://secure.cihi.ca/cihiweb/products/Diabetes_care_gaps_dis parities_aib_e.pdf. Accessed May 1, 2012. 10. Yu N, Su H, Tsai S, et al. ABC control of diabetes: survey data from National Diabetes Health Promotion Centers in Taiwan. Diabetes Res Clin Pract 2009;84: 194e200. 11. Wild S, Roglic G, Green A, Sicree R, King H. Global prevalence of diabetes: estimates for the year 2000 and projections for 2030. Diabetes Care 2004;27: 1047e53. 12. Canadian Diabetes Association. Canadian Diabetes Association 2008 Clinical Practice Guidelines for the Prevention and Management of Diabetes in Canada. Can J Diabetes 2008;32(Suppl 1):S1e201. 13. American Diabetes Association. Standards of Medical Care in Diabetes e 2011. Diabetes Care 2010;34(Suppl 1):S11e61. 14. National Institute for Health and Clinical Excellence (NICE). Type 2 diabetes: The management of type 2 Diabetes 2009. http://www.nice.org.uk/nicemedia/ live/12165/44320/44320.pdf. 15. Manns BJ, Tonelli M, Zhang J, et al. Enrolment in primary care networks: impact on outcomes and processes of care for patients with diabetes. CMAJ 2012;184: E144e52. 16. Halm EA, Lee C, Chassin M. Is volume related to outcome in health care? A systematic review and methodologic critique of the literature. Ann Intern Med 2002;137:511e20.
17. Holmboe ES, Wang Y, Tate JP, Meehan TP. The effects of patient volume on the quality of diabetic care for Medicare beneficiaries. Med Care 2006;44:1073e7. 18. Jamtvedt G, Young JM, Kristoffersen DT, et al. Does telling people what they have been doing change what they do? A systematic review of the effects of audit and feedback. Qual Saf Health Care 2006;15:433e6. 19. Hahn KA, Ferrante JM, Crosson JC, Hudson SV, Crabtree BF. Diabetes flow sheet use associated with guideline adherence. Ann Fam Med 2008;6:235e8. 20. Moharram MM, Farahat FM. Quality improvement of diabetes care using flow sheets in family health practice. Saudi Med J 2008;29:98e101. 21. Ruoff G, Gray LS. Using a flow sheet to improve performance in treatment of elderly patients with type 2 diabetes. Fam Med 1999;31:331e6. 22. Hunt DL, Haynes RB, Hanna SE, Smith K. Effects of computer-based clinical decision support systems on physician performance and patient outcomes: a systematic review. JAMA 1998;280:1339e46. 23. Johnston M, Langton K, Hayes B, Mathieu A. Effects of computer-based clinical decision support systems on clinician performance and patient outcome. A critical appraisal of research. Ann Int Med 1994;120:133e42. 24. Ornstein SM, Garr DR, Jenkins RG, Rust PF, Arnon A. Computer-generated physician and patient reminders. Tools to improve population adherence to selected preventive services. J Fam Pract 1991;32:82e90. 25. Shea S, DuMouchel W, Bahamonde L. A meta-analysis of 16 randomized controlled trials to evaluate computer-based clinical reminder systems for preventive care in the ambulatory setting. J Am Med Inform Assoc 1996;3: 399e409. 26. Coppell KJ, Anderson K, Williams SM, et al. The quality of diabetes care: a comparison between patients enrolled and not enrolled on a regional diabetes register. Primary Care Diabetes 2011;5:131e7. 27. O’Mullane M, McHugh S, Bradley CP. Informing the development of a national diabetes register in Ireland: a literature review of the impact of patient registration on diabetes care. Informat Primary Care 2010;18:157e68. 28. Public Health Agency of Canada. Canadian Adult National Immunization Coverage (Adult NICS) Survey, 2006. http://resources.cpha.ca/CCIAP/data/544e. pdf. Accessed May 1, 2012. 29. Arguedas JA, Perez MI, Wright JM. Treatment blood pressure targets for hypertension. Cochrane Database Syst Rev (Epub July 8, 2009). 30. Cooper-DeHoff RM, Gong Y, Handberg EM, et al. Tight blood pressure control and cardiovascular outcomes among hypertensive patients with diabetes and coronary artery disease. JAMA 2010;304:61e8. 31. ACCORD Study Group, Cushman WC, Evans GW, et al. Effects of intensive blood pressure control in type 2 diabetes. N Engl J Med 2010;362:1575e85. 32. Sui JTP, Tejani AM, Musini V, Bassett K, Mintzes B, Wright J. Hypertension control in patients with diabetes. Can Fam Phys 2011;57:31e3. 33. Genest J, McPherson R, Frohlich J, et al. 2009 Canadian Cardiovascular Society/ Canadian guidelines for the diagnosis and treatment of dyslipidemia and prevention of cardiovascular disease in the adult e 2009 recommendations. Can J Cardiol 2009;25:567e79. 34. Public Health Agency of Canada. Diabetes in Canada 2011. http://www.phacaspc.gc.ca/cd-mc/publications/diabetes-diabete/facts-figures-faits-chiffres-2011/ chap1-eng.php. Accessed July 15, 2013.
Contents lists available at ScienceDirect
Canadian Journal of Diabetes journal homepage: www.canadianjournalofdiabetes.com
Original Research
Quality of Diabetes Care in the Canadian Forces Amole Khadilkar MD a, *, Jeff Whitehead MD a, c, Monica Taljaard PhD a, b, Doug Manuel MD a, b a b c
Department of Epidemiology and Community Medicine, University of Ottawa, Ottawa, Ontario, Canada Clinical Epidemiology Program, Ottawa Hospital Research Institute, Ottawa, Ontario, Canada Department of National Defence, Canadian Forces Health Services, Ottawa, Ontario, Canada
a r t i c l e i n f o
a b s t r a c t
Article history: Received 1 February 2013 Received in revised form 21 July 2013 Accepted 18 August 2013
Background: Published data on quality of care indicators from various countries indicate the challenges of providing high-quality diabetes care. The objective of this study was to evaluate the quality of care provided to members of the Canadian Forces (CF) who have diabetes, by determining the extent to which healthcare providers adhere to recommendations outlined in the 2008 Canadian Diabetes Association (CDA) clinical practice guidelines. Methods: All 14 CF bases meeting eligibility criteria were included in the evaluation. Cases of diabetes were ascertained based on laboratory criteria. Adherence to 21 CDA guideline recommendations was evaluated following a review of patient medical records. Results: The CF demonstrated high adherence (>75%) with 9 recommendations, moderate adherence (50% to 75%) with 7 recommendations and low adherence (<50%) with 5 recommendations. Most notably, there were 4 recommendations for which adherence was greater than 90%. The mean rate of adherence with all applicable recommendations per patient was 60.3% (95% Confidence Interval [CI], 59.0% to 61.6%). CF adherence rates were generally similar to or better than comparable rates in the civilian population within Canada and other industrialized countries. Conclusions: It is unclear what accounts for the favourable quality of diabetes care in the CF Health Services, but this highly structured practice setting has a number of features that distinguish it from provincial healthcare systems. Several strategies can be considered to improve diabetes care even further, including providing feedback to physicians about their performance, promoting the use of diabetes care flow sheets and creating a diabetes registry. Ó 2014 Canadian Diabetes Association
Keywords: quality of care Clinical Practice Guidelines Canadian Forces
r é s u m é Mots clés : qualité des soins lignes directrices de pratique clinique Forces canadiennes
Introduction : Les données publiées sur les indicateurs de la qualité des soins de plusieurs pays montrent les défis liés à la prestation de soins de grande qualité aux diabétiques. L’objectif de cette étude était d’évaluer la qualité des soins fournis aux membres des Forces canadiennes (FC) qui ont le diabète en déterminant dans quelle mesure les prestataires de soins respectent les recommandations définies dans les lignes directrices de pratique clinique 2008 de l’Association canadienne du diabète (ACD). Méthodes : Les 14 bases des FC répondant aux critères d’admissibilité ont été incluses dans l’évaluation. Les cas de diabète ont été établis selon les critères de laboratoire. L’observance des 21 recommandations provenant des lignes directrices de l’ACD a été évaluée d’après une revue de dossiers médicaux. Résultats : Les FC ont démontré une observance élevée (> 75 %) si 9 recommandations avaient été suivies, une observance modérée (50 % à 75 %) si 7 recommandations avaient été suivies et une faible observance (< 50 %) si 5 recommandations avaient été suivies. Plus particulièrement, il y a eu 4 recommandations pour lesquelles l’observance a été plus grande que 90 %. Le taux moyen d’observance de toutes les recommandations applicables par patient a été de 60,3 % (intervalle de confiance [IC] à 95 %, 59,0 % à 61,6 %). Les taux d’observance des FC ont généralement été similaires ou meilleurs que les taux comparables de la population civile de tout le Canada et d’autres pays industrialisés. Conclusions : Il est difficile de savoir ce qui relève d’une qualité de soins favorable dans les Services de santé des FC, mais ce cadre de pratique hautement structuré a de nombreuses caractéristiques qui le distinguent des systèmes de soins de santé provinciaux. Plusieurs stratégies peuvent être considérées
* Address for correspondence: Amole Khadilkar, MD, Department of Epidemiology and Community Medicine, Faculty of Medicine, University of Ottawa, 451 Smyth Road, Room 3105B, Ottawa, Ontario K1H 8M5, Canada. E-mail address: [email protected] 1499-2671/$ e see front matter Ó 2014 Canadian Diabetes Association http://dx.doi.org/10.1016/j.jcjd.2013.08.264
12
A. Khadilkar et al. / Can J Diabetes 38 (2014) 11e16
pour améliorer encore davantage les soins du diabète, y compris l’offre d’une rétroaction aux médecins sur leur performance, la promotion de l’utilisation des organigrammes des soins diabétologiques et la création d’un registre du diabète. Ó 2014 Canadian Diabetes Association
Introduction Between 2008 and 2009, there were an estimated 2.4 million Canadians with diabetes, representing 6.8% of the population (1). Diabetes is the 6th leading cause of death in Canada, with cardiovascular disease representing a major cause of morbidity and mortality among affected individuals (2,3). There is a 2- to 4-fold increased risk of cardiovascular events among people with diabetes compared to those without diabetes, an increase that persists even after accounting for other traditional risk factors (4). Additional complications include blindness, kidney disease, foot ulcers and amputations. The annual cost of diabetes to the Canadian healthcare system is projected to exceed $19 billion by 2020 (5). Indicators of the quality of diabetes care internationally have demonstrated that there are significant challenges to providing high-quality diabetes care (6e10). With the prevalence of diabetes increasing globally due to an aging population and increasing levels of obesity and physical inactivity, the quality of diabetes care has never been a more important issue (11). The objective of this study was to evaluate the quality of care provided to members of the Canadian Forces (CF) who have diabetes, by determining the extent to which health care providers adhere to recommendations outlined in the 2008 Canadian Diabetes Association (CDA) clinical practice guidelines (12).The recommendations proposed in the CDA guidelines were based on the best available evidence as well as expert opinion on best practice. They shared much in common with the American Diabetes Association (ADA) clinical practice guidelines and the UK’s National Institute for Health and Clinical Excellence (NICE) clinical practice guidelines; although there were notable differences as well (13,14). A secondary objective was to compare the quality of diabetes care in the CF with national and international estimates for the general population.
notes or periodic health assessments (diagnostic assessment extract) or who were identified from a pharmacy database as having been prescribed diabetes medication and/or glucose testing strips in the past year. The diagnostic assessment extract contained the diabetes-specific ICD-10 codes (E10-E14) listed for every relevant clinic visit as early as July 2010 up to December 19, 2011. From this initial list, eligible patients for this study were identified by reviewing laboratory records. Patients were included in this study if they had plasma glucose levels in the diabetes range on 2 separate occasions as per clinical practice guidelines (fasting plasma glucose 7.0 mmol/L, random plasma glucose 11.1 mmol/L or 2-hour plasma glucose 11.1 mmol/L after a 75 g oral glucose tolerance test) (12). As the 2008 CDA clinical practice guidelines did not include HbA1c thresholds in their diagnostic criteria, HbA1c levels were not considered. Note that while patients may be prescribed medications, like Metformin for conditions such as pre-diabetes and polycystic ovarian syndrome or require glucose testing strips for monitoring gestational diabetes, the requirement for laboratory evidence of diabetes meant that these conditions were excluded from consideration. Because the CF does not accept applicants with pre-existing diabetes, CF members currently with diabetes must have been diagnosed while in service and would, therefore, have laboratory results corresponding to that diagnosis in their medical records. Quality of care indicators The criteria used to assess quality of care were based on adherence to CDA guideline recommendations in 2 areas: 15 processes of care and 6 intermediate outcomes. Guideline recommendations that should be “considered” were not included in the evaluation as they are open to clinical judgement.
Methods
Process of care measures
A retrospective medical case review of all eligible CF members with diabetes was conducted over a 3-year period to examine adherence to 21 CDA guideline recommendations.
In terms of processes of care, the 2008 CDA guidelines recommended that patients with diabetes should have the following: a hemoglobin A1c test at least every 6 months; an annual test for urine protein excretion (random urine albumin-creatinine ratio [ACR]); an annual kidney function test (serum creatinine converted to an estimated glomerular filtration rate [eGFR]); a blood pressure check at every visit; an annual foot exam with additional testing for peripheral neuropathy; a dilated eye exam every 1 to 2 years; a fasting lipid profile every 1 to 3 years and a baseline EKG in those over 40 years of age (12). An EKG should be repeated every 2 years in those who are at high risk for cardiovascular events (7). Patients at high risk for cardiovascular events were defined as males 45 years old and as females 50 years old (12). All other patients with diabetes were defined as moderate risk. Additional recommendations included an annual influenza vaccination, encouragement to stop smoking, the use of metformin as the initial oral anti-hyperglycemic agent, and the prescription of an angiotensin-converting enzyme (ACE) inhibitor or an angiotensin receptor blocker (ARB) for those with persistent microalbuminuria (ACR 2.0 mg/mmoL in males or 2.8 mg/mmoL in females), even in the absence of hypertension (12). According to the 2008 CDA clinical practice guidelines, an ACE inhibitor or ARB should also be prescribed for those at high risk of cardiovascular events (12).
Study population The study population consisted of members of the CF Regular Force aged 18 to 60 years with diabetes who were located at the 14 bases that incorporated progress notes into the electronic medical record (EMR) for 6 months or more as of December 19th, 2011. There was a total of 32 603 members of the CF Regular Force at these bases, which represents roughly half of the overall CF Regular Force population. The bases included members of the army, navy and air force and were located in all regions of Canada except the Maritime provinces, which did not meet the electronic medical record (EMR) criteria for inclusion, and the territories, where there are no CF bases. Case ascertainment of diabetes CF members with diabetes were identified by extracting a list of patients at the 14 selected bases who either had a diagnosis of diabetes documented in the EMR within the physicians’ progress
A. Khadilkar et al. / Can J Diabetes 38 (2014) 11e16
Intermediate outcomes In terms of intermediate outcomes, the 2008 CDA clinical practice guidelines indicated that among patients with diabetes, hemoglobin A1c levels should be 7.0%, blood pressure should be <130/80 mm Hg and, in those at high risk for cardiovascular events, low-density lipoprotein (LDL) should be 2.0 mmol/L with a secondary target being a total cholesterol:high density lipoprotein (TC:HDL) ratio of less than 4.0 (12). In patients with an ontreatment LDL level of 2.0 to 2.5 mmol/L, clinical judgement should be used to decide if additional therapy is necessary (12). Given that additional therapy for LDL levels between 2.0 mmol/L and 2.5 mmol/L was at the discretion of physicians, they were given the benefit of the doubt and assessed as adhering to guideline recommendations if LDL levels were 2.5 mmol/L. In those patients who were at moderate risk for cardiovascular events, the Canadian dyslipidemia guidelines at the time indicated that cholesterol lowering treatment should be initiated if LDL was >3.5 mmol/L and the TC:HDL ratio was 5.0 (33). Data extraction One abstractor (AK) extracted the data from the medical records. A second abstractor (JW) independently reviewed a random sample of 10% of the charts to determine the inter-rater reliability for 10 of the most important guideline recommendations using the kappa statistic.
Statistical analysis Adherence to each of the 21 guideline recommendations was calculated as proportions with 95% large-sample confidence intervals (CIs). In addition, the mean rate of adherence with the entire panel of applicable recommendations per patient was calculated with 95% CIs. Indirect age and sex standardization of diabetes prevalence rates in the CF was conducted using age and sex-specific Canadian prevalence rates from 2008 to 2009 (34). All statistical analyses were performed using Stata software (version 11.0; StataCorp LP, College Station, TX, USA).
National and international comparisons Results were compared with quality of care data from Alberta’s primary care networks, which were obtained from a large cohort study involving 77464 diabetes patients (15). Primary care networks (PCN) represent a multidisciplinary practice model with a unique funding scheme in which health professionals are paid on a traditional fee-for-service basis and an additional $50 per patient is provided to the PCN to support its ongoing activities (15). Aside from the Canadian Community Health Survey (CCHS), the PCN cohort study is the largest Canadian study reporting on quality of diabetes care. The CCHS had a much more narrow focus with fewer indicators (9). In addition to the PCN quality of care data, comparisons were made with international estimates obtained from the published literature and in the case of the CCHS, the unpublished literature (8e10,15).
Ethics clearance The study was approved by the Director of Access to Information and Privacy (DAIP), as well as Veritas, an independent review ethics board accredited by the Association for the Accreditation of Human Research Protection Programs (AAHRPP).
13
Table 1 Characteristics of the study population Frequency (%)* Gender Males Females Age Mean (SD) Age <40 years Age 40e49 years Age 50e60 years Rank Officers (university education, higher rank, higher income) Noncommissioned members (less education, lower rank, lower income) Smoking status Smokers Nonsmokers Type of diabetes Diet-controlled type 2 diabetes Type 2 diabetes on oral agents Type 2 diabetes on insulin Type 1 diabetes
383 (95.8%) 17 (4.3%) 46.2 53 218 129
(6.3) (13.3%) (54.5%) (32.3%)
45 (11.3%) 355 (88.8%)
100 (25.4%) 294 (74.6%) 32 326 33 9
(8.0%) (81.5%) (8.3%) (2.3%)
SD, standard deviation. * Unless otherwise indicated.
Results The characteristics of the study population are presented in Table 1. A total of 400 diabetes patients were identified using the diagnostic assessment extract and the pharmacy database at the 14 bases included in the study, giving a total diabetes prevalence of 1.2% (95% CI, 1.1e1.3%). One factor contributing to the relatively low diabetes prevalence is the younger population in the CF, two-thirds of who are under the age of 40 years. It is well recognized that the risk of diabetes increases with age. Using the age and sex-specific diabetes prevalence rates in the Canadian population as a reference standard, the total expected number of cases among females in the CF is 106, which is more than the observed number of 17 by a factor of 6 (indirectly standardized prevalence ratio 0.16, 95% CI 0.094e0.26). The total expected number of cases among males is 730, which is more than the observed number of 383 by a factor of almost 2 (indirectly standardized prevalence ratio 0.52, 95% CI 0.47e0.58). Therefore, even after age standardization, diabetes prevalence rates in the CF appear to be lower than that of the general Canadian population. Increased physical activity among members of the CF may be a possible explanation for some of the decreased risk relative to the general population. In addition, the presence of comorbidities among members of the CF with diabetes could be grounds for early medical release, thus reducing the number of people captured in our prevalence statistics. Finally, it is known that women in the CF have significantly lower rates of obesity (17%) compared to their male counterparts (25%), which may account for their particularly lower risk. The level of inter-rater agreement for individual quality of care indicators was high, with the kappa statistic ranging from 0.75 to 0.96. The adherence rates for the 21 CDA guideline recommendations considered in this study are shown in Table 2. For the CF regular force, there were 9 recommendations for which adherence was greater than 75%, including 4 recommendations for which adherence was greater than 90%. These 4 recommendations included using metformin as the initial oral anti-hyperglycemic agent, lipid profile testing every 1 to 3 years, performing a baseline EKG in those over 40 years of age, and prescribing an angiotensinconverting enzyme (ACE) inhibitor or angiotensin receptor blocker (ARB) in patients with persistent microalbuminuria. There
14
A. Khadilkar et al. / Can J Diabetes 38 (2014) 11e16
Table 2 Adherence with the Canadian Diabetes Association clinical practice guideline recommendations among patients within the CF Regular Force and the civilian population (point estimate with 95% confidence interval where available) Recommendation Processes of Care Metformin should be the initial oral anti-hyperglycemic agent prescribed Lipid profile every 1 to 3 years Baseline EKG in patients over 40 years of age ACE inhibitor or ARB prescription in patients with persistent microalbuminuria Smoking cessation advice Eye exam every 1e2 years Annual eGFR HbA1c every 6 months ACE inhibitor or ARB prescription in patients at high risk Annual ACR Follow up EKG every 2 years in patients at high risk Annual influenza vaccination Annual foot exam Annual peripheral neuropathy testing. Blood pressure check at every visit Intermediate outcomes LDL 3.5 mmol/L (moderate risk) TC:HDL < 5.0 (moderate risk) LDL 2.5 mmol/L (high risk) TC:HDL < 4.0 (high risk) HbA1c 7.0% Blood pressure <130/80
CF Regular Force (n¼400)
Alberta’s Primary Care Networks (n¼77464) [15]
Best of International estimates [8e10]
98.9% (97.8e100.0%)
84.3% (83.1e85.5%)
-
96.8% (95.0e98.5%)
63.5% (63.1e63.8%)*
71% to 85% (U.S.)*
96.4% (94.4e98.4%)
-
-
90.9% (82.1e99.8%)
79.3% (78.2e80.3%)
-
82.2% 77.1% 75.6% 75.3% 71.3%
32.7% (32.4e33.0%) * 70.0% (69.7e70.3%)* -
32% to 77% (Australia)y 75% to 85% (U.S.) 65% to 93% (Australia)* -
67.4% (62.8e72.0%) 67.1% (60.9e73.4%)
44.7% (44.4e45.1%) -
-
43.4% (38.5e48.2%) 15.9% (12.3e19.5% 10.3% (7.3e13.3%)
-
51% (Canada) -
7.3% (4.7e9.8%)
-
-
53.2% (52.8e53.6%) -
31% to 47% (U.S.)z 38% to 57% (Australia) 30.9% (Taiwan)
85.3% 74.8% 71.9% 57.5% 53.9% 37.7%
(74.6e89.8%) (73.0e81.3%) (71.4e79.9%) (71.0e79.5%) (65.8e76.9%)
(79.3e91.3%) (67.6e82.0%) (66.4e77.5%) (51.5e63.6%) (48.8e59.0%) (32.7e42.8%)
ACE, angiotensin converting enzyme; ARB, angiotensin receptor blocker; ACR, albumin-creatinine ratio; EKG, electrocardiogram; LDL, low-density lipoprotein; TC:HDL, total cholesterol to high-density lipoprotein ratio; HbA1c, hemoglobin A1c. * Annual estimate. y Biennial estimate. z LDL<2.6 mmol/L in all patients regardless of risk.
was moderate adherence for 7 recommendations with rates ranging between 50% and 75%. There was low adherence (below 50%) for 5 recommendations. The mean adherence with all applicable CDA recommendations per patient was 60.3% (95% CI, 59.0e61.6%). Based on the 7 recommendations for which comparable data were available, the CF generally showed better adherence rates than Alberta’s PCNs. Differences were positive for 6 recommendations ranging from 5.3% to 44.4%; for one recommendation (maintaining a HbA1c at 7.0%), adherence was similar in the 2 populations. Compared to the best of international estimates, the CF showed lower adherence for HbA1c testing, annual kidney function testing and the annual foot exam. However, this study considered a stricter recommendation of HbA1c testing in the last 6 months, whereas Australia and other countries evaluated HbA1c testing in the last year (10). Therefore, corresponding adherence rates would be expected to be higher in those countries. Furthermore, the Australian data that demonstrated higher adherence involved patients visiting specialist diabetes clinics, not primary care practices as was the case in the current study. A higher rate of annual foot exams was found in the Canadian Community Health Survey, which excluded members of the CF, but this represents patient self-reported data that is subject to information bias (9). Furthermore, we only included foot exams documented in the patient’s medical records, which may under-represent performance of the intervention. The guideline adherence rate for the remaining recommendations was the same or better in the CF
compared to countries with the best adherence rates based on international comparisons. Note the CF differs substantially from the populations being compared to in terms of age, gender, socioeconomic status and the presence of comorbidities. However, comparing the quality of care in these different populations based on adherence to the best practices outlined by the CDA has relevance. The CDA guideline recommendations apply equally to people with diabetes of all ages and genders, regardless of the presence of comorbidities and whether the patient is diet controlled or on insulin. Most of the indicators evaluated in this study were process of care measures, which physicians should observe regardless of the health status of the patient. It is acknowledged, however, that older individuals with longer durations of diabetes and multiple comorbidities may be more difficult to manage effectively. Adherence to target intermediate outcomes may be more difficult to achieve in such patients. Therefore, for these specific outcome measures, the comparisons between different populations should be interpreted with caution. Discussion Of the 21 CDA guideline recommendations considered, there was very high (>90%) or high adherence (>75%) for nearly half of the CDA recommendations considered; low adherence (<50%) was observed for only 5 recommendations. Based on comparison of available indicators with Alberta’s primary care networks, the quality of diabetes care in the CF appears to be better than that of
A. Khadilkar et al. / Can J Diabetes 38 (2014) 11e16
15
the civilian population in Canada and better than some international estimates. However, the mean adherence with all applicable CDA recommendations per patient was 60.3%, which is less than optimal. We were rigorous in our application of the CDA recommendations (e.g. adequate HbA1c status required HbA1c testing in the last 6 months). Had we followed a more flexible approach as in many other studies of diabetes care, our indicators would likely have shown an even more favourable picture of diabetes care in the CF. It is not clear what factors related to the Canadian Forces Health Services lend themselves to better diabetes practice. However, physicians caring for members of the CF are salaried, generally see fewer patients per day than physicians working in provincial health care systems, and have more time and resources to provide better comprehensive diabetes care. Medications and medical devices are provided free to CF members, removing financial barriers to treatment; Alberta residents must pay for medications on their own, typically through private insurance. The occupational repercussions of poorly controlled diabetes may motivate CF members to adhere more closely to a physician’s medical advice and engage in appropriate self-management. Finally, the use of the EMR likely plays a role in facilitating better diabetes practice in the CF. On the other hand, there are factors that make diabetes care in the CF more challenging, such as the periodic transfers of CF members to new bases and care teams and the relatively low volume of patients with diabetes (16,17). Despite the fact that the Canadian Forces Health Services demonstrated very good performance in many areas, there is room for improvement. Specific areas where there was low adherence (<50%) included receiving an annual influenza vaccination, maintaining a blood pressure at less than 130/80, getting an annual foot exam, having annual peripheral neuropathy testing and getting a blood pressure check at every medical visit. There are at least 5 quality improvement initiatives that are currently not used in the CF but may be considered to improve the suboptimal areas of diabetes care. First, performing an audit and providing feedback to health care providers about their performance is one useful strategy. Feedback has been shown to improve the quality of patient care, particularly in areas where performance is poor (18). Importantly, the feedback given to physicians should provide encouragement about the areas where they excel, of which several examples are identified in this study. Second, diabetes care flow sheets have been shown to be beneficial in improving guideline adherence, but are used by only a minority of CF physicians (19e21). More widespread use of diabetes care flow sheets would improve the documentation of important processes of care and intermediate outcomes and serve as a convenient reminder system for physicians. Third, clinical decision support tools that provide electronic reminders at the point of care have been shown to improve the quality of care, but they are not supported by the CF’s current EMR software (22e25). Fourth, making 10 g monofilaments available to all physicians for peripheral neuropathy testing is another potential strategy to improve care. Indeed, some CF physicians documented that 10 g monofilaments were not available as a reason for not performing peripheral neuropathy testing. Fifth, creating a diabetes registry may be an important step to further improving the quality of diabetes care in the CF (26,27). This appears feasible based on a case validation performed in conjunction with this study (not reported) and also supports a number of quality of care initiatives including several of the ones noted above.
likelihood of this is low. Second, it could be reasonably argued that the quality of care at the 14 bases selected for this study is not representative of the entire CF. However, these bases constituted almost 50% of the CF Regular Force. Moreover, it is unlikely that bases that had incorporated progress notes into the EMR for 6 months or more differed substantially from bases that did not, as the EMR was sequentially implemented on bases according to factors that were unlikely to be associated with quality of care such as geographic location and convenience. Third, for categorizing adherence to the annual influenza vaccination recommendation, we considered both the current season and the previous season since the current season was not over by the time of the chart review: December 19, 2011. While reasonable, this may have inflated the calculated vaccination rate, which was still well below the 80% target established for Canadians with chronic medical conditions (28). Fourth, in spite of the CDA recommendation to maintain a blood pressure <130/80, there is some controversy about how to define appropriate blood pressures in people with diabetes in light of several recent studies suggesting that intensive blood pressure control is not beneficial (29e32). Therefore, this particular CDA recommendation and the recommendation to have a blood pressure check at every visit may change in future updates of the guidelines. This has implications for the number of suboptimal areas of diabetes care in the CF. Adherence to target intermediate outcomes may be more difficult to achieve in populations with longer histories of diabetes and a higher rate of comorbidities. Therefore, for these specific outcomes, the comparisons between the CF and other populations should be interpreted with caution. Finally, quality of care may be affected by patient age, sex and socioeconomic status. Differences in these socio-demographic variables between the CF and the general population may, therefore, bias comparisons of quality of care between the 2 populations.
Limitations
Dr. Amole Khadilkar contributed substantially to the acquisition, analysis and interpretation of data and drafted the article. All authors contributed substantially to the conception and design of the research, revised the article critically for important intellectual content and gave final approval of the version to be published.
There were several limitations to our study. First, it is possible that we may have missed patients with diet-controlled diabetes who did not visit their physicians often. However, we believe the
Conclusion and future directions In conclusion, the CF fared reasonably well in spite of this rigorous quality of care assessment. Nearly half of the CDA recommendations evaluated had adherence rates over 75%. Moreover, the quality of diabetes care at the 14 bases investigated within the CF was generally similar to or better than the quality of care in the civilian population within Canada and other industrialized countries. Several quality improvement initiatives that may further improve diabetes care in the CF were identified with the creation of a diabetes registry being an important next step. Acknowledgements Special thanks to Major (Dr.) Stephen Cooper for guidance on the use of the EMR, to Robert Hawes for compiling the diagnostic assessment extract, to Janice Ma for providing access to the pharmacy database, to Zelma Buckley for delivering the paper charts and to Laura Bogaert for assistance with Stata programming. Author’s Disclosures There were no conflicts of interest with any of the authors. There were no financial disclosures. Author’s Contributions
16
A. Khadilkar et al. / Can J Diabetes 38 (2014) 11e16
References 1. Public Health Agency of Canada. Diabetes in Canada: Facts and figures from a public health perspective 2011. http://www.phac-aspc.gc.ca/cd-mc/publications/ diabetes-diabete/facts-figures-faits-chiffres-2011/chap1-eng.php. Accessed May 1, 2012. 2. Statistics Canada. Leading Causes of deaths in Canada 2008. http://www.statcan. gc.ca/pub/84-215-x/2011001/table-tableau/tbl001-eng.htm. Accessed May 1, 2012. 3. Miettinen H, Lehto S, Salomaa V, et al. Impact of diabetes on mortality after the first myocardial infarction. The FINMONICA Myocardial Infarction Register Study Group. Diabetes Care 1998;21:69e75. 4. Haffner SM, Lehto S, Rönnemaa T, et al. Mortality from coronary heart disease in subjects with type 2 diabetes and in nondiabetic subjects with and without prior myocardial infarction. N Engl J Med 1998;339:229e34. 5. Canadian Diabetes Association. The prevalence and costs of Diabetes 2008. http://www.diabetes.ca/files/prevalence-and-costs.pdf. Accessed May 1, 2012. 6. McGlynn EA, Asch SM, Adams J, et al. The quality of health care delivered to adults in the United States. N Engl J Med 2003;348:2635e45. 7. Harris SB, Ekoé JM, Zdanowicz Y, Webster-Bogaert S. Glycemic control and morbidity in the Canadian primary care setting (results of the Diabetes in Canada Evaluation study). Diabetes Res Clin Pract 2005;70:90e7. 8. Si D, Bailie R, Wang Z, et al. Comparison of diabetes management in five countries for general and indigenous populations: an internet-based review. BMC Health Services Research 2010;10:169. 9. Canadian Institute for Health Information (CIHI). Diabetes care gaps and disparities 2009. http://secure.cihi.ca/cihiweb/products/Diabetes_care_gaps_dis parities_aib_e.pdf. Accessed May 1, 2012. 10. Yu N, Su H, Tsai S, et al. ABC control of diabetes: survey data from National Diabetes Health Promotion Centers in Taiwan. Diabetes Res Clin Pract 2009;84: 194e200. 11. Wild S, Roglic G, Green A, Sicree R, King H. Global prevalence of diabetes: estimates for the year 2000 and projections for 2030. Diabetes Care 2004;27: 1047e53. 12. Canadian Diabetes Association. Canadian Diabetes Association 2008 Clinical Practice Guidelines for the Prevention and Management of Diabetes in Canada. Can J Diabetes 2008;32(Suppl 1):S1e201. 13. American Diabetes Association. Standards of Medical Care in Diabetes e 2011. Diabetes Care 2010;34(Suppl 1):S11e61. 14. National Institute for Health and Clinical Excellence (NICE). Type 2 diabetes: The management of type 2 Diabetes 2009. http://www.nice.org.uk/nicemedia/ live/12165/44320/44320.pdf. 15. Manns BJ, Tonelli M, Zhang J, et al. Enrolment in primary care networks: impact on outcomes and processes of care for patients with diabetes. CMAJ 2012;184: E144e52. 16. Halm EA, Lee C, Chassin M. Is volume related to outcome in health care? A systematic review and methodologic critique of the literature. Ann Intern Med 2002;137:511e20.
17. Holmboe ES, Wang Y, Tate JP, Meehan TP. The effects of patient volume on the quality of diabetic care for Medicare beneficiaries. Med Care 2006;44:1073e7. 18. Jamtvedt G, Young JM, Kristoffersen DT, et al. Does telling people what they have been doing change what they do? A systematic review of the effects of audit and feedback. Qual Saf Health Care 2006;15:433e6. 19. Hahn KA, Ferrante JM, Crosson JC, Hudson SV, Crabtree BF. Diabetes flow sheet use associated with guideline adherence. Ann Fam Med 2008;6:235e8. 20. Moharram MM, Farahat FM. Quality improvement of diabetes care using flow sheets in family health practice. Saudi Med J 2008;29:98e101. 21. Ruoff G, Gray LS. Using a flow sheet to improve performance in treatment of elderly patients with type 2 diabetes. Fam Med 1999;31:331e6. 22. Hunt DL, Haynes RB, Hanna SE, Smith K. Effects of computer-based clinical decision support systems on physician performance and patient outcomes: a systematic review. JAMA 1998;280:1339e46. 23. Johnston M, Langton K, Hayes B, Mathieu A. Effects of computer-based clinical decision support systems on clinician performance and patient outcome. A critical appraisal of research. Ann Int Med 1994;120:133e42. 24. Ornstein SM, Garr DR, Jenkins RG, Rust PF, Arnon A. Computer-generated physician and patient reminders. Tools to improve population adherence to selected preventive services. J Fam Pract 1991;32:82e90. 25. Shea S, DuMouchel W, Bahamonde L. A meta-analysis of 16 randomized controlled trials to evaluate computer-based clinical reminder systems for preventive care in the ambulatory setting. J Am Med Inform Assoc 1996;3: 399e409. 26. Coppell KJ, Anderson K, Williams SM, et al. The quality of diabetes care: a comparison between patients enrolled and not enrolled on a regional diabetes register. Primary Care Diabetes 2011;5:131e7. 27. O’Mullane M, McHugh S, Bradley CP. Informing the development of a national diabetes register in Ireland: a literature review of the impact of patient registration on diabetes care. Informat Primary Care 2010;18:157e68. 28. Public Health Agency of Canada. Canadian Adult National Immunization Coverage (Adult NICS) Survey, 2006. http://resources.cpha.ca/CCIAP/data/544e. pdf. Accessed May 1, 2012. 29. Arguedas JA, Perez MI, Wright JM. Treatment blood pressure targets for hypertension. Cochrane Database Syst Rev (Epub July 8, 2009). 30. Cooper-DeHoff RM, Gong Y, Handberg EM, et al. Tight blood pressure control and cardiovascular outcomes among hypertensive patients with diabetes and coronary artery disease. JAMA 2010;304:61e8. 31. ACCORD Study Group, Cushman WC, Evans GW, et al. Effects of intensive blood pressure control in type 2 diabetes. N Engl J Med 2010;362:1575e85. 32. Sui JTP, Tejani AM, Musini V, Bassett K, Mintzes B, Wright J. Hypertension control in patients with diabetes. Can Fam Phys 2011;57:31e3. 33. Genest J, McPherson R, Frohlich J, et al. 2009 Canadian Cardiovascular Society/ Canadian guidelines for the diagnosis and treatment of dyslipidemia and prevention of cardiovascular disease in the adult e 2009 recommendations. Can J Cardiol 2009;25:567e79. 34. Public Health Agency of Canada. Diabetes in Canada 2011. http://www.phacaspc.gc.ca/cd-mc/publications/diabetes-diabete/facts-figures-faits-chiffres-2011/ chap1-eng.php. Accessed July 15, 2013.