The only way to cope with more chronic disease patients with better quality of treatment in the face of limited personnel and funds is to use IT to transform the practice into a more efficient one.
Discussions regarding the quality of healthcare today mostly focus on the availability of qualified personnel and effective and safe drugs. Qualified personnel are scarce and their professional education is expensive and time consuming. The cost and time factor is true for the development of effective and safe pharmaceutical treatment also. This situation has led to (and rightly so) to measures being taken to utilise precious resources in the most efficient way possible. The two most important areas the healthcare and life sciences industry has focused on, in this respect, are business process reengineering to organise the business/organisation/practice more efficiently and installation of intelligent software solutions, which should both support the more efficient organization, as well as the quality of outcomes.
Software solutions are less expensive and more dependable than people and drugs, and therefore, it may be expected that apart from improvements in efficiency and quality of healthcare, it may be expected that this should be possible at a limited additional cost.
Case study: The Netherlands
In The Netherlands several dedicated healthcare software development companies have been founded to contribute to this important, though ambitious goal; and this region also serves as the test-bed for implementation of the new software. These companies meticulously try to develop software that will support the goal of improving efficiency and quality of outcomes at acceptable costs. One amongst these enterprises is VitalHealth Software, founded by the Mayo Clinic of the United States and a social venturing fund of a Dutch IT entrepreneur in The Netherlands. This company has been the subject of a prospective comparative study in order to investigate the clinical benefits and cost-effectiveness of such a software supported approach. VitalHelath Software, acquired for this reason the leading disease management company in The Netherlands, Diagnosis4Health, who had been developing and implementing practice and clinical support software for over 10 years. It was investigated whether the combination of the practical clinical experience and the superior Information Technology of both companies would be able to deliver the promise of a software support solution that will make the necessary contribution to the healthcare system.
Prospective comparative study of usefulness of IT
This project appears to be more than just a dream. This has been demonstrated by the first reports from the reputed Julius Center of the University Medical Center, Utrecht in The Netherlands. In a prospective comparative parallel study, a group of 55 matched practices were randomly organised into two groups, of which the active group (29 practices with 1699 patients) would get the opportunity to use the practice and clinical support software related to care for patients suffering from diabetes type 2 and the control group (22 practices with 1692 patients) would continue with practice as usual regarding the care for a similar patient set. The control group would get to work with the software one year after the start of the study. This pilot test, about which has been published in several journals1, 2, studied not only clinical outcomes, but uated the cost-effectiveness also.
The clinical investigators FGW Cleveringa, M van den Donk, KJ Gorter and GEHM Rutten from the Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, were supported by the health economics experts LW Niessen and WK Redekop from the Institute for Medical Technology Assessment, Erasmus Medical Center, Rotterdam, to assess the clinical outcomes in terms of their effect on the patients lives and to translate these effects into costs for the community. This analysis was conducted with the objective to prove whether or not practice and clinical support software would improve the health condition in patients with a chronic disease in this case diabetes type 2 and whether this improvement could be achieved at an acceptable cost. The outcomes would help in the discussion whether or not this type of software should indeed be applied in daily medical practice, and whether or not we could close the discussion on the subject by rigorously implementing it or by forgetting it for good.
In the randomised trials, it was already shown and published that the so-called Diabetes Care Protocol (DCP, also known as Vital for Diabetes) is able to reduce the cardiovascular risk of type 2 diabetes patients.1 The DCP intervention consists of an organisational aspect, a software tool and repeated reporting. The organisational aspect means that a diabetes consultation hour is set up and run by a qualified practice nurse. The software tool consists of a computerised decision support system, a recall system and three monthly feed-back information.
In both groups patients were seen at baseline and after 1-year for their annual diabetes check-up. Among other things HbA1c, blood pressure, cholesterol, medication, diabetic complications were registered. Data of individual patients in both groups after a year were used to compute average lifetime disease outcomes, direct medical costs and quality adjusted life years (QALY) per patient. A validated probabilistic Dutch diabetes model and the UKPDS risk engine were used to extrapolate the results.2 This model included health states for cardiovascular disease (CV) and other severe DM2 related complications (blindness, end stage renal disease, lower extremity amputation). Costs of medication use and implementing and maintaining DCP were included. Incremental costs per QALY were calculated. Since 50% of the population had a history of cardiovascular disease (CV+), the model assumes for 60-64 year old patients with CV+ an increased risk of a cardiovascular event (i.e., relative risk of 1.47 in men and 2.95 in women).
Table 1 summarises the effectiveness results from the cluster-randomised trial. There were statistically significant differences in cardiovascular parameters at the end of the 1-year follow-up. These differences resulted in a reduction in 10-year UKPDS coronary heart disease risk estimates.
Table 2 shows the lifetime QALYs and costs for DCP and control treatment. Patients in the DCP group live 0.63 years longer than patients in the control group, and experience 0.38 more QALYs. Patients in the DCP group incur slightly higher medical costs than patients in the control group (EUR 130). These differences (EUR 130, 0.38 QALYs) result in an incremental cost-effectiveness ratio of EUR 342 per QALY gained.
DCP would cost about EUR 300-400 per patient over a lifetime. However, it would reduce the risk of CV events and thereby reduce CV event costs.
Table 1. DCP trial outcomes of clinical parameters within and between groups
(N = 3391)
* The 10 year United Kingdom Prospective Diabetes Study (UKPDS) coronary heart disease (CHD) risk (%) was calculated using: date of onset of diabetes (age duration of diabetes), sex, ethnicity, smoking, HbA1c, systolic blood pressure, total cholesterol and HDL-cholesterol
generalised linear model
for percentages the OR is given
Improvements of intervention group compared to control group significant (p<0.05)
Table 2. Costs and effects of the DCP compared to usual care: Base-case results
One-year results show that DCP is improving the quality of outcomes and reduces cardiovascular risk in patients with diabetes type 2, resulting in a extension of life of 0,63 years of which 0,38 years are spent in good quality (0,38 QALY’s). With costs of EUR 342 per QALY gained DCP is very cost-effective as compared to usual care in The Netherlands.
Based upon these results health authorities, health insurers and medical professionals are increasingly interested in implementing this type of IT supported care systems. In The Netherlands this is expected to lead to structural improvement of the quality of medical care. In countries where healthcare has not yet reached the quality standard of The Netherlands, the DCP approach offers a big chance to make tremendous developments towards quality care. The other countries do not have to make huge investments now in R&D, and they don’t have to go through the long, painful and expensive journey of trial and error. They can take advantage of these results, minus the original investment and with a much shorter time-to-market, which may just be very suitable for countries, which are in the process of their development.
The author is grateful to Dr. Frits Cleveringa c.s. for allowing him to cite from his published data and presentations.
Cleveringa et al. Combined task delegation, computerised decision support and feedback improve cardiovascular risk for type 2 diabetes patients. A cluster randomised trial in primary care. Diabetes Care 2008;10.2337/dc08-0312.
Dijkstra et al. A Patient-centred and professional-directed implementation strategies for diabetes guidelines: a cluster-randomised trial-based cost-effectiveness analysis. Diabet Med. 2006 Feb;23(2):164-70.