The Ontario Osteoporosis Strategy recently released its technical report which I encourage all family physicians, endocrinologists and orthopaedic surgeons to read. The report takes data from ICES and CIHI and brings to light some interesting province-wide and LHIN-wide trends in osteoporosis screening.
The main takeaway for me from the report is that the current system is doing a generally poor job of screening patients for osteoporosis. Top level points include:
- In 2014/15 only 13% of men and 28% of women aged 68-70 had received a BMD in the past 5 years
- In 2014/15, there was a three-fold difference between high and low rates for standardized BMD testing following a hip fracture (low of 11.9 in Erie St. Clair; high of 38.6 in Toronto Central)
It goes without saying, but there is an awful lot of improvement when you look at that table. Many eligible patients not being screened at all, and many not receiving appropriate care post-fracture.
But before every LHIN jumps out of their seat wanting to improve this as a quality metric, we need to step back and practically evaluate which groups would truly benefit from screening.
The 2010 Canadian clinical practice guidelines recommend screening all individuals >= 65 years old (men and women) for osteoporosis. The Osteoporosis Self-Assessment Tool is a validated screen that is useful in determining which patients would warrant a BMD. The formula is Weight (kg) – Age (years), and if <10, increased risk of osteoporosis and BMD is warranted.
How do these two approaches to screening compare in a real-life setting? I looked at my practice-level data and will share the insights here.
In a practice of 2200 patients, 243 women >= 65yo were identified, 224 of whom have had at least one lifetime BMD (92%). 250 men >= 65yo were identified, 107 of whom have had at least one lifetime BMD (43%).
l went through my patient data and analyzed which patients have a calculated OST < 10 and would qualify for BMD screening, sorted by age cohort.
First, the men (11 patients were excluded because of no recorded weight):
65-69yo: 9/75 have an OST <10 (12%)
70-74yo: 24/61 (39%)
75-79yo: 23/37 (62%)
80-84yo: 26/31 (84%)
>=85yo: 31/31 (100%)
And the women (21 patients were excluded because of no recorded weight):
65-69yo: 34/52 have an OST <10 (65%)
70-74yo: 38/59 (64%)
75-79yo: 29/38 (76%)
80-84yo: 32/33 (97%)
>=85yo: 47/47 (100%)
In total, 45% of men >=65yo would qualify to be screened with BMD according to the OST, and 77% of women >=65yo would qualify.
Certainly looking at the male data, it appears that using the age 65 cut off is a very crude screen and may lead to over screening with BMD. But what about the most important outcome, which men actually required treatment for osteoporosis as a consequence of screening those >=65yo? (using Mayo Clinic resource as shared decision-making tool)
I identified 14 male patients in my practice who are currently being treatment for osteoporosis (either with bisphosphonate or denosumab): 10 patients >=65yo, 6 patients 50-64yo, and 1 patient <50yo. Interestingly, only one of the 14 patients was identified as having osteoporosis through the >=65yo screening program. The remainder were identified as requiring a BMD because of other risk factors:
– 5 with compression fractures (2 of whom have alcohol use disorder)
– 2 on prednisone
– 2 with rheumatoid arthritis
– 1 with Parkinson’s
– 2 with alcohol use disorder
– 1 with Prader-Wili
Out of 107 men >=65yo who received at least one lifetime BMD, only ONE patient lacking any other risk factors apart from age warranted (and agreed to) treatment for osteoporosis.
A few points to consider:
- If we are going to target osteoporosis screening as a quality metric to improve, I think it behooves us to isolate screening in women >=65yo as the actionable, measurable metric, rather than including both men and women in the cohort. Screening all men >= 65yo will not be a cost-effective intervention, as it will not identify a significant number of men who will actually warrant treatment, and will instead result in many completely fruitless BMDs, at a cost of millions of dollars to the system. While likely not measurable as a widespread metric, the preferred approach to screening in men should be risk factor-based, with a particular focus on screening men with a history of corticosteroid use, alcohol use disorder, RA, those on aromatase inhibitors, and those with a previous fragility fracture (and screening men with chronic back pain with plain x-ray to rule out compression fracture).
- The OST should be encouraged as the initial screening evaluation prior to BMD, and that it be highlighted to physicians that 100% screening compliance is not the suggested target in women >=65yo (only 77% were eligible according to OST in my patient population).
- The technical report mentioned that of those on treatment for osteoporosis, 38% of individuals received a follow-up BMD within 2 years. While the report frames this as a proportion that should be increased, I would take the opposite approach to this data. Ordering a repeat BMD within the first 2 years of treatment for osteoporosis is rarely indicated as the results should not often change treatment. Any change in the BMD within 2 years of starting treatment is just as likely to be inherent variability in the test as compared to a meaningful change due to treatment. We should be targeting that 0% of patients receive a follow-up BMD within 2 years of starting treatment.
Certainly there is tremendous room for improvement in how we are systematically screening for osteoporosis in Ontario, especially in those who have already experienced a fragility fracture and in women >=65yo. We must, however, also ensure that any widespread quality improvement efforts must be rational in who it targets and careful to avoid over screening in extremely low-risk populations, such as men who lack any risk factors other than age.