I think he might be using data for instance from this non-access article: http://www.ncbi.nlm.nih.gov/pubmed... A total of 20/1518 patients and 1/1733 controls carried the G2019S mutation. Specificity was 99.9% (95% CI, 99.6-100%), sensitivity was 1.3% (0.8-2.1%), and the positive likelihood ratio was 22.8. A positive family history of PD increased the positive likelihood ratio to 82.5. - Attila Csordas

And here is how I'd calculate this 22.8 LR based only on those data using that only 1 carried the mutation out of the controls: (1733x20)/1518=22.8326746... Am I right or wrong? I haven't used the sensitivity, specificity data & forgot what I learned on this at the uni. Any expert here available to confirm or not? - Attila Csordas

the whole Brin sentence: "In fact, it is somewhere between 20% to 80% depending on the study and how you measure." So based on the article we have around 20% (22.8) PLR without knowing the positive family history and circa 80% (82.5) with a positive family history of PD. - Attila Csordas

another data on G2019S: haplotype analysis in 504 cases with PD and 314 controls enrolled in the Genetic Epidemiology of PD Study. Cases and controls were recruited without knowledge of family history of PD and cases were oversampled in the < or =50 age at onset (AAO) category. RESULTS: The LRRK2 G2019S mutation was present in 28 cases with PD (5.6%) and two controls (0.6%) (chi(2) = 13.25; p < 0.01; odds ratio 9.18, 95% CI: 2.17 to 38.8). The mutations L1114L, I1122V, R1441C, and Y1699C were not identified. The frequency of the LRRK2 G2019S mutation was 4.9% in 245 cases with AAO < or =50 years vs 6.2% in 259 cases with AAO >50 (p = 0.56). All cases with PD with the G2019S mutation shared the same disease-associated haplotype. - Attila Csordas