There is no evidence that the current or historical screening programmes for DDH have improved the functional outcomes of adult patients with DDH partly because there has never been a clear gold standard definition of what constitutes a case of DDH.

Rates of surgery have been used as a surrogate endpoint for the effectiveness of a screening programme with the assumption that early diagnosis and treatment of DDH will allow stabilisation through conservative measures, reducing the requirement for surgical management at a later date. However, as the indications for surgical intervention are not universally agreed, results may be affected by changing thresholds for surgery over a period of time, or in different geographic regions.

Similarly, the incidence of “late presenting” DDH in a screened population has been used as an endpoint, where late presentation represents a failure of screening: again, there is significant variability in the definition of ‘late’ and in how long the follow up is. Observational studies, assume a “captive population” or at least one in which children moving into the geographic area are of similar number and have the same rates of late presentation to those leaving the area.

Finally, rates of splintage/treatment have been used to judge the success of surveillance programmes based on the assumption that an increased splintage rate should be reflected in a subsequent decrease in the rates of late presenting DDH. Historically, some programmes saw an increased treatment rate with little reduction in late presentation [26, 27], raising concerns about a high false positive rate and over treatment, with a potential risk of resultant AVN. More recent trials have gone on to demonstrate the opposite [28, 29], perhaps reflecting improved implementation of screening programmes.

There are no studies directly comparing an unscreened cohort with that of a cohort undergoing a clinical screening programme alone. Table 3.2 lists a number of observational studies that have reported the rates of late presenting DDH in different patient populations. These studies suggest that the effectiveness of a clinical screening programme is improved when the person performing the examination has experience. Hadlow [30] demonstrated reductions in the rates of late presentation of DDH when a Consultant Orthopaedic surgeon, rather than a junior paediatrician performed clinical examination. These rates remained low when experienced paediatricians performed the examinations in subsequent years. This work was supported by similar studies from Krikler [31] and Tegnander et al. [32]. Interestingly, Moore hypothesised that the act of examination itself may cause instability through forceful or repeated attempts to dislocate an otherwise stable hip [33]. There was no evidence in this paper (and none since) to support this hypothesis.

The use of ultrasound as part of surveillance for DDH is well established, with certain centres electing to use universal surveillance of newborns. A number of observational studies have compared the introduction of universal surveillance to that of historic controls in order to estimate the effect on rates of late presentation and splintage [27, 28, 37–40]. These demonstrate a variable increase in splintage rates, with improvement in the rates of late presenting DDH, with one study stating that no cases have presented late since the introduction of universal screening [39]. There are presently only two controlled trials (level II evidence) comparing the techniques directly.

Two arms of the 1994 Rosendahl et al. [41] controlled trial compared clinical examination alone to clinical examination and universal ultrasound in neonates. The study reported increased rates of splintage in the universal ultrasound group without a statistically significant reduction in the rates of late diagnosed DDH (Table 3.3). **Even with the large numbers used in this study, due to the low rates of late presentation, later power calculations suggested that the original study was underpowered to demonstrate significance in late diagnosis [42].

In 2002, Elbourne [29] compared the use of ultrasound vs serial clinical examination in patients with clinical instability as part of a multicentre randomised controlled trial. The study demonstrated lower rates of splintage in hips, which were initially clinically abnormal, when ultrasound was used as part of the assessment. There was no difference seen in later surgery rates, suggesting that the ultrasound group was not undertreated.

Ultrasound assessment programmes can either be universal or selective with the latter aiming to identify those patients at highest risk of DDH. Risk factors for DDH have been well described, with a recent meta-analysis demonstrating a significant increase in relative risk for female infants, breech presentation, first born and those with a positive family history [43]. Historically, a foot deformity has often been considered to have an association with DDH, but the meta-analysis in 2012 showed this did not reach statistical significance [44]. Selective screening programmes reserve ultrasound examination for those infants who have an abnormal clinical examination and/or one or more risk factors. The most frequently used risk factors are a positive family history, breech positioning and foot deformities [41, 45–49]: female gender is not considered a risk factor for screening programmes.

Two large controlled trials (Table 3.4) have compared selective ultrasound screening with universal ultrasound [41, 48]. Neither demonstrated statistically significant differences in the rates of late diagnosis between the two patient groups. In Holen’s study [48], 5 patients were diagnosed late in the selective group (all without risk factors) and one patient from the universal group (where the protocol was not adhered to and no ultrasound actually took place). The authors of both trials [41, 48] emphasise that their ultrasound surveillance programmes supplemented a clinical programme where the examination was performed by an experienced physician, suggesting that their results are dependent on a combination of clinical and ultrasound assessments. Inexperienced clinical examination may lead to a loss of patients from the selective ultrasound screening group [31, 32] and a higher late diagnosis rate.