An update of the statistical methods underlying the FELS method of skeletal maturity assessment

Division of Morphological Sciences and Biostatistics, Lifespan Health Research Center, Department of Community Health and.
Annals of Human Biology (Impact Factor: 1.27). 09/2013; 40(6). DOI: 10.3109/03014460.2013.806591
Source: PubMed

ABSTRACT Abstract Background: Evaluation of skeletal maturity provides clinicians and researchers a window into the developmental progress of the skeleton. The FELS method for maturity assessment provides a point estimate and standard error based on 98 skeletal indicators. Aim: This paper outlines the statistical methodology used by the original FELS method and evaluates improvements that address the following: serial correlation in the calibration sample is now considered, a Bayesian estimation method is now employed to improve estimation near ages 0 and 18 years and uncertainty in the calibration due to sampling is now accounted for when computing confidence limits. Subjects and methods: The original FELS method was calibrated using 677 Fels Longitudinal Study participants. In the improved method, serial correlation is accounted for using GEE, a Bayesian analysis with a prior centred on chronological age is used and the bootstrap is used to account for all sources of variation. Results: Accounting for serial correlation resulted in larger slopes for ordinal indicators. The Bayesian paradigm led to narrower confidence limits and a natural interpretation of skeletal age. Sampling variability in the calibration parameters was negligible. Conclusion: Improvements to the statistical basis of the FELS method provide a more effective method of estimating skeletal maturity.

1 Follower
19 Reads
  • Source
    • "The method most frequently employed for skeletal age assessment is the atlas of Greulich and Pyle (GP) [16]. Moreover, the so-called bone-specific techniques are widespread: the method of Tanner Whitehouse (TW) in the latest versions TW2 [17] and TW3 [18]; and the method of Roche et al. [19] (FELS method), which is utilized and studied primarily for clinical purposes [20] [21]. Finally, in some German-speaking countries, the Thiemann HH, Nitz I. atlas method [22] is used. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Paediatricians, radiologists, anthropologists and medico-legal specialists are often called as experts in order to provide age estimation (AE) for forensic purposes. The literature recommends performing the X-rays of the left hand and wrist (HW-XR) for skeletal age estimation. The method most frequently employed is the Greulich and Pyle (GP) method. In addition, the so-called bone-specific techniques are also applied including the method of Tanner Whitehouse (TW) in the latest versions TW2 and TW3. To compare skeletal age and chronological age in a large sample of children and adolescents using GP, TW2 and TW3 methods in order to establish which of these is the most reliable for forensic purposes. The sample consisted of 307 HW-XRs of Italian children or adolescents, 145 females and 162 males aged between 6 and 20 years. The radiographies were scored according to the GP, TW2RUS and TW3RUS methods by one investigator. The results' reliability was assessed using intraclass correlation coefficient. Wilcoxon signed-rank test and Student t-test were performed to search for significant differences between skeletal and chronological ages. The distributions of the differences between estimated and chronological age, by means of boxplots, show how median differences for TW3 and GP methods are generally very close to 0. Hypothesis tests' results were obtained, with respect to the sex, both for the entire group of individuals and people grouped by age. Results show no significant differences among estimated and chronological age for TW3 and, to a lesser extent, GP. The TW2 proved to be the worst of the three methods. Our results support the conclusion that the TW2 method is not reliable for AE for forensic purpose. The GP and TW3 methods have proved to be reliable in males. For females, the best method was found to be TW3. When performing forensic age estimation in subjects around 14 years of age, it could be advisable to use and associate the TW3 and GP methods.
    Forensic science international 03/2014; 238C:83-90. DOI:10.1016/j.forsciint.2014.02.030 · 2.14 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Skeletal maturity assessment provides information on a child's physical development and expectations based on chronological age. Given recently recognized trends for earlier maturity in a variety of systems, most notably puberty, examination of sex-specific secular trends in skeletal maturation is important. For the orthopaedist, recent trends and changes in developmental timing can affect clinical management (eg, treatment timing) if they are currently based on outdated sources. (1) Has the male or female pediatric skeleton experienced a secular trend for earlier maturation over the past 80 years? (2) Do all indicators of maturity trend in the same direction (earlier versus later)? In this retrospective study, a total of 1240 children were examined longitudinally through hand-wrist radiographs for skeletal maturity based on the Fels method. All subjects participate in the Fels Longitudinal Study based in Ohio and were born between 1930 and 1964 for the "early" cohort and between 1965 and 2001 for the "recent" cohort. Sex-specific secular trends were estimated for (1) mean relative skeletal maturity through linear mixed models; and (2) median age of maturation for individual maturity indicators through logistic regression and generalized estimating equations. Overall relative skeletal maturity was significantly advanced in the recent cohort (maximum difference of 5 months at age 13 years for girls, 4 months at age 15 years for boys). For individual maturity indicators, the direction and magnitude of secular trends varied by indicator type and sex. The following statistically significant secular trends were found: (1) earlier maturation of indicators of fusion in both sexes (4 months for girls, 3 months for boys); (2) later maturation of indicators of projection in long bones in both sexes (3 months for girls, 2 months for boys); (3) earlier maturation of indicators of density (4 months) and projection (3 months) in carpals and density in long bones (6 months), for girls only; and (4) later maturation of indicators of long bone shape (3 months) for boys only. A secular trend has occurred in the tempo of maturation of individual components of the pediatric skeleton, and it has occurred in a sex-specific manner. The mosaic nature of this trend, with both earlier and later maturation of individual components of the skeletal age phenotype, calls for greater attention to specific aspects of maturation in addition to the overall skeletal age estimate. The Fels method is currently the most robust method for capturing these components, and future work by our group will deliver an updated, user-friendly version of the Fels assessment tool. Appreciation of sex-specific secular changes in maturation is important for clinical management, including treatment timing, of orthopaedic patients, because children today exhibit a different pattern of maturation than children on whom original maturity assessments were based (including Fels and Greulich-Pyle).
    Clinical Orthopaedics and Related Research 02/2015; 473(8). DOI:10.1007/s11999-015-4213-1 · 2.77 Impact Factor