Compliance, or more correctly concordance, is the key issue in osteoporosis treatment. Concordance is the preferred term as it rightly implies that therapeutic choices are the result of shared decision making. However, whatever term we use, if patients do not take their medicine they have no chance of benefiting from it.
The recent guidance from the National Osteoporosis Guideline Group (NOGG) on prevention and treatment of osteoporosis covers the assessment and diagnosis of both men and women.1 In this article, I have focused on the following questions in relation to this guideline:
- how bad is patient compliance with bisphosphonates?
- does the NOGG guidance offer a realistic way to stratify risk?
- why did the guideline group stop short of offering algorithms that could be embedded into GP computer systems?
Concordance with bisphosphonates
Most patients who are prescribed bisphosphonates do not take their medicine as prescribed. In our quality improvement work in this area we found that less than half of women aged over 45 years who were prescribed bisphosphonates actually collected prescriptions for the drugs.2 However, this is likely to be an underestimate of the number of people actually taking their bisphosphonate because some people who do collect their prescription will not take it to the pharmacy. Of those who do go to the pharmacy, not everyone will actually take their medicine.3
My PCT (Guildford and Waverley) ran a programme called COUNT to identify people not likely to be taking their medicines.3 The mnemonic ‘COUNT’ stands for: Confused, Over-ordering, Unable to open, Not taking medicines, Too many journeys.
The service consisted of a pharmacist visit to the surgery to check the records of those patients who had a high COUNT score; and then a visit to the patient’s home to give them advice in person. Bisphosphonates were one of the most common medicines ‘not taken’.
Gastrointestinal side-effects are probably the main reason that patients do not comply with their bisphosphonate treatment—in a recent systematic review, the common side-effects listed across nearly all studies of this class of medicine were:4
- abdominal pain
- upper gastrointestinal symptoms
To overcome these problems with gastrointestinal side-effects, drug manufacturers have developed preparations with increasingly longer intervals between doses. The first change was to extend the dosage interval from daily to 1 week. Although this did improve concordance, it still remained poor.5 More recently we have seen the arrival of quarterly (ibandronic acid) and yearly intravenous (zoledronic acid) bisphosphonates, although their place in the therapeutic armamentarium is yet to be finalised.
Stratifying risk—NOGG versus NICE?
The NOGG guidance makes fracture risk assessment straightforward if you use the computer interface available at www.shef.ac.uk/FRAX. It also addresses what to do if therapy is not tolerated. You simply enter the patient’s data in the fracture risk assessment (FRAX®) calculator (see Figure 1) and then click on the ‘View NOGG guidance’ box in the lower part of the risk calculator. This takes you to a screen that gives a visual display of the fracture risk with some guidance text underneath on classification of the patient as low, intermediate, or high risk. Figure 2 shows the assessment and treatment thresholds from which fracture probability can be computed. The patient detailed in Figure 1 (aged 76 with a 10-year probability of major osteoporotic fracture of 34%) would therefore be at high-risk according to Figure 2.
The displayed recommendations regarding treatment are simple and easy to follow. Alendronate is recommended as first-line treatment in the majority of cases. It is a low-cost drug with a broad spectrum of anti-fracture efficacy. Other bisphosphonates, raloxifene, or strontium ranelate may be prescribed if a patient is intolerant of alendronate. Parathyroid hormone peptides are expensive and, therefore, are only recommended for use in those at very high risk of fracture.1
There is also a paper version of the FRAX® tool, but the healthcare professional would have to pull together all the risk factors and interpret them in the context of the patient’s body mass index. This method may take longer and would be more prone to errors.
The equivalent NICE guidance is much harder to find and interpret as it is spread across various documents—you may need to look at several guidelines to find the advice needed.6,7 Once you find the right piece of NICE guidance, it is easy to use—the quick reference guide which explains how to treat post-menopausal women who have fragility fractures, is clear and easy to follow. As in the NOGG guidance, the NICE guideline looks at the patient’s age and a number of risk factors, but it also requires the ‘T-score’ to decide whether treatment is required, and only applies to women who have had a previous fracture.8
The single NOGG computer interface, which can be used for nearly everybody (i.e. those in the specified age range for whom there is an evidence-base) is clearly advantageous as it is user-friendly and much quicker to use.
Figure 1: Example data entry using the FRAX® tool
Reproduced with kind permission from the WHO Collaborating Centre for Metabolic Bone Diseases www.shef.ac.uk/FRAX/
Figure 2: Assessment and treatment thresholds without (left) or with (right) a BMD test to compute fracture probability for men and women
Figure taken from: Kanis J, McCloskey E, Johansson H et al; and the National Osteoporosis Guideline Group. Case finding for the management of osteoporosis with FRAX®–Assessment and intervention thresholds for the UK. Osteoporos Int 2008; 19 (10): 1395–1408.
Why has the NOGG guideline not been embedded into practice computer systems?
Although the NOGG guideline is useful, it clearly has limitations:
- it can only be used for one patient at a time and not for audit
- it requires the entry of data already contained in the GP computer system
- it does not identify the large number of non-compliant people who are at high risk of fracture.
Therefore, embedding the NOGG guidance into GP computer systems would be useful and more efficient for the user. The duplicate data entry required to use FRAX® online is time consuming and can lead to transcription errors. Much of the data required for the risk score is already well recorded in GP computer systems, for example: age, gender, weight, height, smoking status, steroid dose, rheumatoid arthritis, and alcohol consumption. Other factors are less well recorded, such as fractures, parental hip fracture, secondary osteoporosis, and T-score.
Missing data will usually result in risk being underestimated. High-risk patients are still worth detecting even if, for some, there is an underestimate of risk. Recording of T-score is a particular problem and may have to be excluded as one leading brand of GP computer system does not allow the entry of negative numbers.9
Embedding the NOGG guidance and FRAX® calculator into the GP computer system would also allow opportunistic identification of patients at high-risk of osteoporotic fracture and those who are non-compliant. It would also allow the identification of those individuals not collecting their bisphosphonate prescriptions to be flagged-up as part of a regular audit. The availability of such a tool would allow the high-risk group who are either not being treated or who are non-compliant, to be identified in the same way that people with chronic kidney disease have been flagged using practice records.10
The National Osteoporosis Guideline Group should be congratulated on developing a tool that provides a single online site, which can be used to determine fracture risk and whether there is a need for drug therapy. It is a tool that can feasibly be used in the 10-minute patient consultation.
However, the group should not rest on its laurels for too long as improvements can still be made. The algorithms that underpin this risk calculator could be used to audit GP records regularly for high-risk patients who are not collecting their prescriptions. The focus on improving osteoporosis management should be on identifying and supporting those people who are not taking their drug treatment.
- National Osteoporosis Guideline Group on behalf of the Bone Research Society, British Geriatrics Society, British Orthopaedic Association, British Society of Rheumatology, National Osteoporosis Society, Osteoporosis 2000, Osteoporosis Dorset, Primary Care Rheumatology Society, and Society for Endocrinology. Osteoporosis—clinical guideline for prevention and treatment. London: NOGG, 2008.
- de Lusignan S, van Vlymen J, Hague N, Dhoul N. Using computers to identify non-compliant people at increased risk of osteoporotic fractures in general practice: a cross-sectional study. Osteoporos Int 2006; 17 (12): 1808–1814.
- Guildford and Waverley PCT. Case Study: COUNT. NHS Innovations South East 2006.
- Brandão C, Lima M, da Silva A et al. Treatment of postmenopausal osteoporosis in women: a systematic review. Cad Saude Publica 2008; 24 (4): 592–606.
- Cramer J, Gold D, Silverman S, Lewiecki E. A systematic review of persistence and compliance with bisphosphonates for osteoporosis. Osteoporos Int 2007; 18 (8): 1023–1031.
- National Institute for Health and Care Excellence. Alendronate, etidronate, risedronate, raloxifene, strontium ranelate and teriparatide for the primary prevention of osteoporotic fragility fractures in postmenopausal women. Technology Appraisal 160. London: NICE, 2008.
- National Institute for Health and Care Excellence. Alendronate, etidronate, risedronate, raloxifene, strontium ranelate and teriparatide for the secondary prevention of osteoporotic fragility fractures in postmenopausal women. Technology Appraisal 161. London: NICE, 2008.
- National Institute for Health and Care Excellence. Alendronate, etidronate, risedronate, raloxifene, strontium ranelate and teriparatide for the secondary prevention of osteoporotic fragility fractures in postmenopausal women. Technology Appraisal 161. Quick Reference Guide. London: NICE, 2008.
- de Lusignan S, Valentin T, Chan T et al. Problems with primary care data quality: osteoporosis as an exemplar. Inform Prim Care 2004; 12 (3): 147–156.
- de Lusignan S, Chan T, Stevens P et al. Identifying patients with chronic kidney disease from general practice computer records. Fam Pract 2005; 22 (3): 234–241.G