“Research highlights” is a weekly round-up of research papers appearing in the print BMJ. We start off with this week’s research questions, before providing more detail on some individual research papers and accompanying articles.

• Could screening for breast cancer cause more harm than good?
• What is the extent of overdiagnosis resulting from detection of non-progressive cancer by screening mammography?
• Does starting antihypertensive treatment in people aged 80 or over reduce cardiovascular events quickly enough to support starting treatment?
• Can bisphosphonate use improve implant survival after total arthroplasty of the lower limb?

Weighing the benefits and harms of mammography screening
The controversy over the value of mammography screening for breast cancer has continued for years, with a Cochrane review in 2009 concluding: “It is thus not clear whether screening does more good than harm.” This week, two modelling studies attempt to clarify the balance of benefits and harms from screening.

James Raftery and Maria Chorozoglou have updated the life table analysis of the 1986 Forrest report (which led to the introduction of the UK breast screening programme) and combined harms and benefits of screening into a single measure (quality adjusted life years (QALYs)). They found negative net QALYs in the early years after the start of screening (up to 10 years after), after which positive net QALYs accrued, but at a slower rate than reported in the original Forrest report.

The main potential harms of mammography screening are overdiagnosis and overtreatment (detection and removal of lesions that would not have progressed to invasive cancer), and Arnaud Seigneurin and colleagues designed a stochastic simulation model to estimate the degree of overdiagnosis among women aged 50-70 in a region of France. They calculated that overdiagnosis accounted for only 1.5% of all cases of invasive cancer and for 28% of all cases of carcinoma in situ (though carcinoma in situ accounted for less than 15% of all incident cases of breast cancer). They conclude that overdiagnosis of invasive cancers was smaller than expected—especially compared with studies that looked at incidence rates before and after implementation of breast screening, which have reported 30-50% overdiagnosis for invasive cancers.

However, as the accompanying editorial by Allan Hackshaw points out, the results of such modelling studies depend on the reliability of the parameters used and the underlying assumptions of the model. Readers’ responses to the full studies published on bmj.com show just how controversial these can be.

Do hip and knee replacements last longer with bisphosphonates?
Another study generating many responses online is Daniel Prieto-Alhambra and colleagues’ cohort study of bisphosphonates in patients who’ve undergone total arthroplasty of the knee or hip. These operations are very common, but so is subsequent revision surgery—as the authors say in the full paper online, about one in 75 patients needed a revision of their prosthesis within three years in 2003—and loosening of the implant, which occurs when the bone supporting it is resorbed, is the most common cause. Bisphosphonates, which reduce bone resorption, are therefore of interest as a potential way to prolong the life of a hip or knee implant, reducing the need for revision surgery.

The retrospective study took data from the United Kingdom’s General Practice Research Database on patients followed up for a median of 3.5 years after primary hip or knee replacement, and concluded that treatment with bisphosphonates did indeed show a strong association with implants lasting longer. However, as the authors state, randomised trials will be needed to confirm such an association, as this study was observational, although the analysis was adjusted for several confounding factors. Responses to the paper online reflect further on possible confounding factors in the study, as well as questioning whether the long term adverse effects of bisphosphonates might outweigh the apparent short term benefits, and speculating on whether the time at which patients receive bisphosphonates is important to the association with implant survival.

Influence of experience on performance of individual surgeons in thyroid surgery
In this prospective cross sectional study by Antoine Duclos and colleagues, patients were at increased risk of permanent complications after a thyroidectomy when operated on by inexperienced surgeons or those who had spent the longest time in practice since graduation. Surgeons aged 35-50 years provided the safest care.

Effects of glucagon-like peptide-1 receptor agonists on weight loss
Treatment with glucagon-like peptide-1 receptor agonists for at least 20 weeks leads to weight loss in obese or overweight patients with or without type 2 diabetes mellitus, according to results of a meta-analysis by Tina Vilsbøll and colleagues. The effect of the drugs might be more pronounced in patients without diabetes, and they also reduce systolic and diastolic blood pressure and total cholesterol, say the authors.