Abstract
Background: Despite many quality improvement trials, diabetes care often remains suboptimal. Few studies in a primary care setting have investigated the ‘real life’ association between organizational differences and quality of diabetes care. Methods: Observational study among ten health care centres with a total of 45 general practitioners (GP). We investigated health care organization and related this to quality of care in a total of 1849 electronic patient records. Results: There were large differences among health care centres in the percentage of patients receiving optimal care (range: 8–67%). The odds to receive good quality of care was higher if the health care centre had a diabetes education program (OR: 4.3; CI: 3.4–5.4), when yearly medical check-ups were done by both the GP and nurse practitioner (NP) (OR: 5.5; CI: 4.2–7.3), planned that after the patient visited the NP the patient is discussed with the GP (OR: 1.8; CI: 1.6–2.0), and had structured follow-up measures for compliance to check-ups (OR: 0.7; CI: 0.5–0.9 and OR: 0.59; CI: 0.5–0.7 for respectively one and two active measures compared to three active measures).
Conclusion: Also in real life, quality of care for type 2 diabetic patients is related to health care organization.
Introduction
The importance of good diabetes care is well known and much research has been done to improve the quality of diabetes care (Citation1–5). Since quality of care is often insufficient too many diabetics do not have a correct metabolic regulation, resulting in increased risk of morbidity and mortality. Therefore, guidelines emphasise the importance of good process of care such as annual haemoglobin A1c determination. Evidence for the effectiveness of quality of care improvement strategies mainly comes from randomised clinical trials. These studies have shown that interventions such as task delegation to a nurse practitioner (NP), reminders and education, have a positive influence on the quality of diabetes care (Citation6–12)
It is important to know how much of the evidence from clinical trials is implemented in everyday practice and which interventions are still effective under these conditions. Few studies have looked at the effect of different organizational aspects in ‘real life practice’ (Citation13–15). Two studies found that organizational aspects such as sending out reminders for check-ups and diabetes self-management education, could improve the quality of care (Citation13,Citation14). A study by Keating et al. found that only a small proportion (5%) of the variation in quality of care could be explained by differences in organizational aspects (Citation15).
More information about ‘real life’ efficacy is needed to understand fully the merits of different organizational aspects. This may help health care organizations to decide which strategies they should implement. The aim of this study, therefore, was to investigate whether differences in quality of care among general practices could be explained by differences in how they organize their diabetes care.
Methods
Setting
In the Netherlands 70–80% of diabetes care takes place outside the hospital. We collected data from ten academic health care centres with a total of 45 general practitioners (GPs) in the south of the Netherlands. These practices are supervised by an umbrella organization called RNH (Registration Network General Practitioners) through visitation and accreditation. All practices are intensively involved in medical education and research by linkage to the Maastricht University (Citation4). All practices use electronic patient files and are used to register medical information.
Patients
All patients with the diagnosis diabetes mellitus type 2 at 1 January 2006, who had one year of subsequent follow-up were included. The patients were identified by the diabetes registration code (flag) and the International Classification of Primary Care (ICPC). We excluded patients who had explicitly refused regular examinations or who were primarily under control of a nursing home physician or endocrinologist.
We gathered patient characteristics such as gender, age, date of diagnosis and relevant co-morbidity. If the date of diagnosis was not correctly documented, we used the date of the first day on which anti-diabetic medication was prescribed or the date of abnormal blood glucose in the laboratory results.
Data collection
Quality of diabetes care. The Dutch diabetes guideline determines how often different diabetes related examinations should be performed. In each diabetic patient we assessed the compliance to the following eight rules:
glycated haemoglobin level (HbA1c) (1/year)
serum LDL cholesterol (LDL) (1/year)
plasma creatinine level (1/year)
proteinuria (1/year)
systolic blood pressure (SBP) (1/year)
weight (3/year)
foot examination (1/year) and
eye examination (1/2 years).
The overall compliance to the rules was determined by one composite score (quality of care (QoC) index), thus ranging from zero to eight.
Organization of diabetes care. Although most health care centres use the services of more than one GP, the diabetes care is usually organized uniformly among the GPs in one centre. This is mainly due to the fact that a nurse practitioner works for more than one GP. Therefore, a questionnaire about the way the diabetes care was organized was administered to one GP and one NP of each primary health care centre. We focussed our questionnaire on those aspects that could discriminate among health care centres and that could potentially be changed (see ). In case of discrepancy between answers of GP and NP of the same health care centre, we considered the answer of the person who was responsible for that specific item decisive.
Appendix 1: Questionnaire about organization of diabetes care, given to one GP and NP of each primary health care centre
Statistical analyses
All analyses were performed with SPSS 14.0 and SAS 9.1. To test the univariate association between organizational aspects and the percentage of patients who received optimal care (i.e. QoC index score >7) chi-square tests were used. Multilevel multiple logistic regression analysis with correction for several patient characteristics was used to determine the isolated effect of the different organizational aspects (independent variable) on quality of care (dependent variable). Multilevel statistics was used because some independent variables (age, co-morbidity) and the dependent variable (quality of care) were measured at patient level, while organizational aspects could only be determined on health care centre level. By using multilevel statistics we were able to adjust for the correlation that existed within health care centres as regards quality of care they provided. In addition, we adjusted for age and gender and for those patient characteristics that were univariately related (logistic regression) to quality of care. A P-value <0.05 was considered to be statistically significant.
Results
Patients
A total of 2743 diabetic patients were potentially eligible for our study. After applying the in- and exclusion criteria, 1849 patients remained for analysis (see ). They had a mean age of 67 years, a mean duration of diabetes of six years, used oral anti diabetics most and had coronary heart disease as most frequent co-morbidity (see ).
Table I. Baseline characteristics of the study population
Figure 1. Selection procedure patient population. HCC, health care centre
Quality of diabetes care
In the percentage of patients receiving optimal care for each health care centre is shown. Large differences among practices were observed. For example, the percentage of patients receiving annual proteinuria measurements in health care centre 3 was 27%, while this was 85% in health care centre 10. Also great variation existed in adherence to the eight rules. Most of the centres measured blood pressure yearly (92%), but many patients (63%) did not receive annual proteinuria measurements. It appears that our practices generally do better than other Dutch centres.
Table II. Percentages of patients of the different health care centres (HCCs) who received diabetes care according to the Dutch guidelines for general practice.
The QoC-indexes per centre are shown in . This figure shows that centres two and five score above 50%. This means that in these centres more than 50% of the patients receive optimal care.
Figure 2. Percentage of patients per health care centre (HCC) with an optimal quality of care index (QoC-index = 8) in 2006.
Association of organizational aspects with quality of diabetes care
shows the simple association between organizational aspects and quality of care. Most of the organizational aspects seemed to be associated with a higher quality of care.
Table III. Association of organizational aspects of diabetes care given to patients with the optimal quality of care index (QoC-index = 8).
The multilevel logistic regression analysis showed that the odds for a patient to receive optimal quality of care was higher if the health care centre had an education program about the disease to her diabetic patients (OR: 4.29; CI: 3.40–5.41), had their yearly medical check-ups done by both the GP and nurse practitioner (OR: 5.51; CI: 4.16–7.30), planned that after the patient visited the NP the patient is discussed with the GP (OR: 1.80 CI: 1.62–2.00), and had made arrangements to ensure that every patient shows up for check-ups (OR: 1.00 for three active measures versus OR: 0.65, CI: 0.47–0.89 and OR: 0.59, CI: 0.50–0.69 for respectively one and two active measures). The call-up and registration system, the protocol for diabetes care and the refresher course lose their significance in our study population using the multivariate model. The use of report cards and multidisciplinary collaboration had no significant effect on the QoC-index.
Several items of the questionnaire were not distinctive, meaning that most centres answered the same. Therefore these items were not used in the analysis. This concerned the questions about three monthly check-ups, insulin regimes, population of diabetics seen by NP and diabetes consulting hour.
Discussion
In this study we analysed the quality of care in relation to organizational aspects of diabetes care in primary care. We observed considerable differences among health care centres in the percentage of patients receiving optimal care. The odds of receiving optimal care is higher when practices have a diabetes education programme, planned that after the patient visited the NP the patient is discussed with the GP, have yearly check ups performed by GP and NP and take active measures in case patients do not show up for check-ups.
Due to our observational design we were able to gain information about ‘real life’ efficacy of several organizational aspects. In addition, because of this study design, we were able to examine several interventions at the same time. Systematic reviews have shown that well-designed observational studies report similar findings as randomised trials (Citation16). Being able to control for potential confounders is crucial in that case. That is why we adjusted for differences in gender and age distribution, and other patient characteristics that were univariately associated with QoC.
One major difference between an RCT and our study, is that in an RCT the health care centres would not be able to choose whether they would implement the interventions. Health care centres that have already adopted several measures to deliver better diabetic care could be the centres that are always the first to adopt new strategies for optimal care. However, being an innovative centre alone probably does not independently predict good quality of diabetes care without taking the active measures that we investigated.
There is much literature on the interventions that we analysed. However, only few of these studies looked at quality of care in an observational setting using a composite score.
Our findings concerning the relation between organizational aspects and quality of care are consistent with two previous studies, which also found associations between organizational aspects of diabetes care and quality of care (Citation13,Citation14). Fleming et al. found statistically significant correlations between quality of care and for example having diabetes self-management education and having opportunities for any practitioner input (Citation14). However, after adjusting for structural and geographical variables, only the practitioner input remained significant (Citation14). Mangione et al. found that greater intensity of structured care management was significantly associated with a higher quality of care (Citation11). They analysed three disease management strategies (e.g. diabetes care management, which included use of patient reminders and use of diabetes education) and found strong associations between the intensity of the strategies and the quality of care (Citation11). However, Keating et al. found that practice management only had a limited non-significant correlation with the quality of diabetes care and that it accounted for only a small proportion of the variation of quality of care (Citation15). Quality of care was non-significantly higher for practices that used quality performance reports, diabetes-specific reports and routinely enrolled diabetic patients in disease management programmes. However, their population was smaller than our study population and the variation of quality of care was low.
disease and teaches skills that are required for self management of diabetes care.
Our finding that a diabetes education program correlated significantly with a higher quality of diabetes care, could be explained by the assumption that patients who are well informed are more aware of the importance of a good metabolic regulation and regular check-ups more. Another important finding of this study is the relevance of structural cooperation between the GP and the NP. If the NP discusses every patient with the GP, they can discuss ways to improve the care. In addition, the chance of ‘forgetting’ measures of diabetes regulation is smaller. When looking at the number of active measures taken in case a patient does not show up for a check up, it seems quite logical that more active measures is related to better quality of care.
Some of the aspects that we studied did not show to be significantly correlated with quality of care. One should take into consideration, however, that this finding could be related to our study population. We performed our study in a Dutch academic environment, and it is possible that quality of care is generally better there. This could result in differences in room for improvement when our population is compared to a population where diabetes care is less well organized. Some of the less strong predictors, as observed in our study, could therefore, become significant predictors in other populations.
In conclusion, our study shows that also in ‘real life’ there is a relationship between organization of diabetes care and quality of diabetes care. Therefore, it seems worthwhile for health care centres to improve the organization of diabetes care.
Ethical approval
All patients gave permission to use their medical data for scientific research when they joined the RNH practices. No further ethical approval was required for this research.
Conflict of interests
The authors declare that they have no conflict of interest with the contents of this paper.
Source of funding
There was no external source of funding for this study.
Related Research Data
References
- Girach A, Manner D, Porta M. Diabetic microvascular complications: Can patients at risk be identified? A review. Int J Clin Pract. 2006;60:1471–83.
- Stratton IM, Adler AI, Neil HA, Matthews DR, Manley SE, Cull CA, . Association of glycaemia with macrovascu-lar and microvascular complications of type 2 diabetes (UKPDS 35): Prospective observational study. Br Med J. 2000;321:405–12.
- Solomon CG. Reducing cardiovascular risk in type 2 diabetes. N Engl J Med. 2003;348:457–9.
- Van den Akker M, Buntinx F, Metsemakers JF, Roos S, Knottnerus JA. Multimorbidity in general practice: prevalence, incidence, and determinants of co-occurring chronic and recurrent diseases. J Clin Epidemiol. 1998;51:367–75.
- Turner RC, Millns H, Neil HA, Stratton IM, Manley SE, Matthews DR, . Risk factors for coronary artery disease in non-insulin dependent diabetes mellitus: United Kingdom prospective diabetes study (UKPDS: 23). Br Med J. 1998; 316:823–8.
- Ubink Veltmaat LJ, Bilo HJ, Groenier KH, Rischen RO, Meyboom de Jong B. Shared care with task delegation to nurses for type 2 diabetes: prospective observational study. Neth J Med. 2005;63:103–10.
- Renders CM, Valk GD, Griffin SJ, Wagner EH, Eijk Van JT, Assendelft WJ. Interventions to improve the management of diabetes in primary care, outpatient, and community settings: A systematic review. Diabetes Care 2001;24:1821–33.
- Stern E, Benbassat CA, Goldfracht M. Impact of a two-arm educational program for improving diabetes care in primary care centres. Int J Clin Pract. 2005;59:1126–30.
- Nyman MA, Murphy ME, Schryver PG, Naessens JM, Smith SA. Improving performance in diabetes care: a multi-component intervention. Eff Clin Pract. 2000;3:205–12.
- Glasgow RE, Nutting PA, King DK, Nelson CC, Cutter G, Gaglio B, . Randomized effectiveness trial of a computer-assisted intervention to improve diabetes care. Diabetes Care 2005;28:33–9.
- Ettner SL, Thompson TJ, Stevens MR, Mangione CM, Kim C, Neil Steers W, . Are physician reimbursement strategies associated with processes of care and patient satisfaction for patients with diabetes in managed care? Health Serv Res. 2006;41:1221–41.
- O’Connor PJ, Crain AL, Rush WA, Sperl Hillen JM, Gutenkauf JJ, Duncan JE. Impact of an electronic medical record on diabetes quality of care. Ann Fam Med. 2005;3:300–6.
- Mangione CM, Gerzoff RB, Williamson DF, Steers WN, Kerr EA, Brown AF, . The association between quality of care and the intensity of diabetes disease management programs. Ann Intern Med. 2006;145:107–16.
- Fleming B, Silver A, Ocepek Welikson K, Keller D. The relationship between organizational systems and clinical quality in diabetes care. Am J Manag Care. 2004;10:934–44.
- Keating NL, Landrum MB, Landon BE, Ayanian JZ, Borbas C, Wolf R, . The influence of physicians’ practice management strategies and financial arrangements on quality of care among patients with diabetes. Med Care 2004;42:829–39.
- Concato J, Shah N, Horwitz RI. Randomized, controlled trials, observational studies, and the hierarchy of research designs. N Engl J Med. 2000;342:1887–92.
- Dijkstra R, Bouma M. Werken aan kwaliteit in de huisartsen-praktijk [Working at quality in general practice] Chapter 4: Diabetes Mellitus. Houten: Bohn Stafleu Van Loghum; 2005.