Your question is like : Is research beneficial for someone? Meta-analysis is one method among other research methods that normally gives you a little more statistical power. Read: https://en.wikipedia.org/wiki/Meta-analysis

Tack för det - yes, Statistical Power. But what is it good for when I am ill? So, my question is not about research in general, but "meta-analysis" of clinical study as a quantitative research synthesis.

Det beror ju helt på vad du studerar. Vad har du för sjukdom? Vill du ha bestrålning, kemoterapi eller operation om du har en liten cancerknöl av en viss sort? Jag antar att du vill ha den metod som har minsta recidivfrekvensen.

Das hängt ganz davon ab, was Sie studieren. Was haben Sie für eine Krankheit? Möchten Sie Strahlung, Chemotherapie oder Operation, wenn Sie eine kleine Krebsart einer bestimmten Sorte haben? Ich nehme an, Sie wollen die Methode mit der niedrigsten Rezidivrate, gell:)

The women were glad when they discovered that the risks were the same when they just took a small part of the breast and did not remove the whole breast relative to ca mammae in Finland.

"den metod som har minsta recidivfrekvensen" -- vid hundra tusen patienter från 2000 året till 2015 året, eller för mig i morgon? (Ursäktar, mitt Svensk är inte bra.) Den är mitt problemet.

Für Sie, jetzt heute. Entschuldigung, jag ser att ni är statistiker och medicinare, ni är säkert bättre på att bedöma meta-analysen eventuella värde i så fall.

Meta-analysis is a statistical technique for combining data from multiple studies on a particular topic. To make a valid decision about using an intervention, ideally we should not rely on the results obtained from single studies. This is because results can vary from one study to another for various reasons, including confounding factors, and the different study samples used.By combining individual studies, and thus using more data, the precision and accuracy of the estimates in the individual studies can be improved upon. Additionally, if the individual studies were under powered, combining them in a meta-analysis can increase the overall statistical power to detect an effect.

How is a meta-analysis performed?

Below are the basic steps involved in a meta-analysis (3):

1- Identifying/formulating a problem (i.e. a question to be answered e.g. to determine the effectiveness of exercise for depression compared with no treatment and comparator treatments). 2- Doing a literature search: this will probably involve searching multiple databases that index reliable peer-reviewed articles, such as PubMed, Scopus, Web of science, Embase, etc. 3- Deciding on selection/inclusion criteria: you should use inclusion and exclusion criteria that will ensure that high-quality evidence, of direct relevance to your research question, is included. For this reason, we tend mostly to include randomized controlled trials (and may exclude observational studies). Ideally, we would also include unpublished studies in order to avoid publication bias. (If we fail to include all of the relevant studies, our conclusions may be erroneous. Specifically, we may overstate the benefit of a treatment (for example), because studies which fail to find a significant effect are less likely to be published than those which do not find a significant effect. See here for more information). 4- Data extraction: you ought to extract data for your outcomes of interest to be pooled (combined) in the final analysis set. 5- Doing the basic meta-analysis: there are a range of software for this purpose, such as Review Manager and Comprehensive Meta-Analysis Software.

Thank you, your account is definitely valid - internally! From a technical point of view, this is a very concise description and useful to understand how to do it, and to understand the analyst's motivation to recommend it to the researcher in the field.

But what about "external validity"? What is the logic by which I can infer from 0.1 - 10 million historical patients to my outcome from tomorrow's surgery? To my knowledge, this link exists by a common unquestioned assumption, for convenience then, only.

Kära Beatrice, jag inte har "i så fall" - men jag hade, och därefter jag tänkte på mig: "What is this meta-analysis 'business' after all, is it good for me? Or, my family?" Så, när är det bra för oss? Då och då? Strängt taget ... när?

Många hälsningar, Jochen

P.S. To your appended information "The women were glad when they discovered that the risks were the same when they just took a small part of the breast and did not remove the whole breast relative to ca mammae in Finland." Assuming, the outcome had been the other way, would the women have been equally happy to know the then-better alternative.(

Yes, to the drug developer and the researcher_They inform of an "aggregate" position of results and how the next research should or can shift such results

Yes, Andrew, This is exactly what I understood when I had worked in industry, as well. It is an answer to requirements by agencies or by research units who have to decide about what way to go: they need a research synthesis, best quantitative. It is a very legitimate method in that context. A different context is clinical patient management, when the issue is how to manage a single patient. Thank you for pointing it out. Best , Jochen

Dear Jochen,

They used to take the whole breast and nobody was happy, then the new result came. It is better to eat a part of the cake than the whole cake.

Dear Béatrice, you are explaining to me that only the MA provided the statistical power needed to demonstrate a higher frequency of "good outcome" with mini-surgical removal of knots (lumpectomy) in comparison with mastectomy. Yes, I understand this very elucidating case. Consequently, with regard to a future patient, the relative odds for good outcome from optional lumpectomy will be taken to exceed the odds for good outcome after optional mastectomy, and the patient will opt for lumpectomy - after the doctor reassured her that her risk profile matches the risk profile of the good-outcome-when-randomized-to-lumpectomy group of patients in the MA.

But I assume, there will also be a group of good-outcome-when-randomized-to-mastectomy patients in the MA study cohorts; then the previous preference may be reviewed. 100 percent success is very rarely seen, and if you do see it, it will hardly have needed a meta-analysis, I assume.

Those women with mastectomy were much more shaken after the experience than those with lumpectomy. If you were looking at yourself after a mastectomy you were recommended to sit down in front of a mirror and have a glas of water to drink when you looked at your mutilated body. Though one woman's husband said that the men always came to terms without breast. Those with so call cake piece or lumpectomy were never suggested to do anything of this sort because it was considered a less traumatizing thing. This happened on a university hospital 5 km from my home. My interest is psychological and to see how certain events affect certain people. If the change in operation method was based on a meta-analysis or not, I cannot say.

Thanks for explaining the situation - and I can now imagine the traumatizing experience better. The trauma in the head (the feeling of a threat on ... what? ... the relationship with the partner?) may be more important in person's anticipation of outcome (prognosis) than the physical trauma; however, the life threat will be realized and come to foreground with exacerbation of disease. Reminds me of a strong sure comrade in his twenties, who, in his end thirties, shot himself after learning that his grandma had AD - may call it "AD-preventive suicide"?

I am addressing "Comprehensive Understanding of (Human) Health and Impact from Person's Surroundings" in a dedicated conference 10-12 December 2018 in Düsseldorf; and I have been quite a lot in touch with those of your colleague-psychologists who sail under "neuro-psychology", "neuroergonomics", "neuro-decision making", "neuro-economics", "neuro-marketing", mixing with neuro-engineering for "cognitive neuro-dynamics". In my thoughts, the body is a whole, only the scientists are split up and kept in their cages (disciplines), and the human body system must then be addressed as a whole. Inside the body, I am working on the complex design of internal controls (biokybernetik), behaviour is outside-world's perception of a person's operations that emerge from within-body functional components' interaction. A mathematical theory can now link the physiology to sociology - in theory.

If this should incidentally be of any interest, you may like to see / keep up with

(1) https://www.researchgate.net/project/GeneSEES-GeneSEES-chuangshi-Comprehensive-Understanding-of-Human-Health-with-Impact-from-Persons-Surroundings ,

(2) https://www.researchgate.net/project/Biokybernetik-engl-bio-automation-russkij-biokibernetika-zhongwen-shengwukongzhilun ,

respectively.

The method of meta-analysis for research in clinical (!, i.e. about patients) medicine will have to get its role in this train of thought. I am currently not sure, where.

Best, Jochen