The "Hygiene Hypothesis”

When selecting a good orange in your market’s produce section, you want to rely on a range of criteria or measurements: size, firmness, a smell that indicates freshness, and, if your market offers samples, taste. In the same way, a scientific article can be subjected to various metrics to gauge its impact, beyond the comparatively simple measure of “times cited.”

InCites, the Clarivate Analytics resource for performance evaluation and benchmarking, provides this capacity. One of the available metrics, Category Normalized Citation Impact (CNCI), affords a closer, more nuanced look at a paper’s impact, compared to other reports published in the same subject area and the same year. This measure can confirm that, even if a paper’s citation count is relatively low, the report has wielded substantial and lasting impact within its particular field.

To illustrate this, we can turn to a 1989 paper in the British Medical Journal.

In the BMJ report, “Hay fever, hygiene, and household size,” David P. Strachan of the London School of Hygiene & Tropical Medicine proposes a theory about the modern rise in the incidence of hay fever, asthma, and other ailments. After examining British records of childhood health from a half-century ago, Strachan offers a seemingly paradoxical hypothesis: modern improvements in hygiene and personal cleanliness, along with smaller households, may have contributed to the recent rise of respiratory ailments and other conditions. In past decades, households with more children, and fewer of today’s hygienic amenities, produced more instances of inter-family infection (older siblings to younger ones, for instance) and the subsequent development of immunity. By comparison, today’s smaller, cleaner households lack those “advantages” in protecting against allergies and other conditions.  

We can demonstrate that Strachan’s 1989 article remains the most important in articulating this “hygiene hypothesis, and how its impact has stood up to more recent reports. In this regard, the CNCI metric is useful in measuring both older and newer reports and how they compare in impact against a specific baseline.  



CNCI can convey a dimension of impact that might not be immediately evident. There are many highly cited articles, but by comparing within the same subject area and same publication year, CNCI evaluates an article’s impact in a normalized context. The above graph, with Strachan’s paper represented by the strikingly tall line, shows that the 1989 report has wielded exceptionally high impact. If we start with the approximately 3,900 articles indexed in Web of Science since the year 1900 on the topic of “hay fever,” we can import those records to InCites Benchmarking for analysis by CNCI. Several articles have amassed more than 1,000 citations each, but their CNCI scores are actually low compared against the high baseline figures for their respective subject fields and years.

The graph demonstrates that many articles have been published on the topic of “hay fever.” Many of them have proposed explanations for the condition’s mechanism, as well as its cure. Notably, however, the 27-year-old theory proposed by Strachan is still attracting citations and is evaluated very highly in the Web of Science category of “Medicine, General & Internal.”