In PLOS ONE, Nov 2017 issue, Chambers and colleagues presented an interesting topic – a systematic review of grandparents’ influence on grandchildren’s cancer risk factors as manifest in the domains of weight, diet, physical activity, smoking use/exposure, and alcohol. This is a very useful resource for researchers and clinicians who work with multi-generational households or with grandparents who are involved in the caretaking their grandchildren.
While their results that grandparents have an adverse effect on cancer risk in their grandchildren is important, it is somewhat difficult to interpret the significance of their findings without further clarification on two major points:

(1) There is no clear description of what criteria the authors used to categorize levels of grandparents impact on their grandchildren. In other words, exactly how did they determine whether the results of a particular study showed a “beneficial,” “adverse,” or “mixed” influence, or “no” impact? For example, Watanabe et al (2011) reported that three-generation family households were significantly associated with a lower risk of irregular mealtimes, and that irregular mealtimes were associated with children being overweight/obese. Chambers et al characterized the findings of Watanabe et al regarding grandparents influence on grandchildren’s cancer risk (specifically on diet) as being “adverse/no impact,” and it is not clear from reading Watanabe et al how Chambers et al made that determination.

(2) The definition of “primary caregiver of grandchild” was not specified precisely. In their methodology section, Chambers et al indicate that they have excluded from their systematic review any studies where grandparents were considered to be “primary caregivers.” However, their review did include studies such as Li et al (2014) where parents answered affirmative to a query of “… whether grandparents play a major role in caring for children (by asking who the child spent most of their time with after school during the weekdays), irrespective of whether or not they live in the household (p.701).” Without further clarification, it is not clear how Chambers et al distinguish a “primary caregiver” from a grandparent who “spends most of their time with…”

(3) There were also two relatively minor points which could use clarification in Table 1.
[a] Sata et al (2015) reported that boys were more likely to be overweight at ages 6 and 12, and added that they are more likely to be overweight even at age 22. However, the finding about age 22 was missing in Chambers et al’s summary. [b] Chambers et al (2017) indicated in Pulgaron et al’s systematic review included 6 studies on the impact of grandparent involvement on child weight.”(p.9). However, Chambers et al cited only 5 of those 6 studies that had been included in Pulgaron et al. It is not clear why the one study from Pulgaron et al was excluded in Chambers et al’s citations.