Part of what’s exciting about reviewing hiker survey data is seeing how actual facts about trail experiences match up to assumptions and accepted ways of doing things. So far, writing Ounces posts has been a lot of fun for this very reason.
Perhaps one of the most ingrained bits of outdoor “common sense” is to always treat water taken from backcountry sources. But is this advice supported by fact?
10 years ago, in the journal Wilderness and Environmental Medicine, Dr. Thomas Welch, MD wrote an editorial in which he challenged conventional wisdom around treating backcountry water. Amongst his remarks, Dr. Welch noted that “…there is no good epidemiologic evidence that North American wilderness waters are inherently unsafe for consumption.” In support of his position, Dr. Welch cited a 2003 Backpacker Magazine article that sampled backcountry water and found the highest concentration of Giardia lamblia cysts to be 1.5 cysts per liter. “Even if one happened upon this spot,” Dr. Welch noted, “it would take nearly 7 L of water consumption to achieve the minimum infective dose of this organism.” He concluded that emphasis on good hand sanitation techniques to stop hand-to-mouth spread is key in eliminating gastrointestinal maladies in the backcountry. Focus on water quality and “routine universal water treatment” should be de-emphasized.
Do the results of the 2013-14 long distance hiker survey support Dr. Welch’s position?
879 long distance hikers who hiked more than 500 miles in a single season on the Appalachian or Pacific Crest Trails answered the following questions:
- Most of the time, did you use something to make your water safe to drink? (yes/no)
- Did trail water ever make you sick on this journey? (yes/no/I don’t know)
In completing this analysis, hikers were divided into two groups, those who treated their water (WT) and those who did not treat their water (NWT). The 78 hikers who answered “I don’t know” to the latter question were removed from the analysis. There was no difference between groups in the proportion of hikers answering “I don’t know.” (Chi-Square, p=0.181). For the remaining 801 hikers, the group distributions were as follows:
|Sick = no||Sick = yes|
|Water treatment = no||100||9|
|Water treatment = yes||628||64|
Based on these numbers, the rate of water sickness in the WT group was 9.25%. The rate of water sickness in the NWT group was 8.26%. There was no significant difference between groups in the number of people who said they experienced sickness due to trail water. (Chi-Square, p=0.738) In other words, using something to make water safe to drink had no effect on the risk of a hiker reporting that they got sick from water.
There are two competing and compelling explanations for this result. The first is that certain water treatment technologies do a better job than others in preventing water sickness, either because they are superior technologies, they are easier to use correctly, or both. The argument follows that no difference was found in the previous analysis because it lumped all treatment technologies together. This argument would “hold water” with the finding that one water treatment technology significantly exceeded the others in rates of water sickness. The chart below sorts the numbers of hikers who got sick from water by the predominant water treatment technology used:
|Sick = no||Sick = yes||Sickness Rate|
|Chlorine Dioxide (tabs or drops)||153||12||7.3%|
On visual inspection, it would seem that users of certain technologies indeed reported higher rates of water sickness than others, but when scrutinized with statistical analysis, no significant difference was found in rates of perceived water sickness between technologies (Chi Square, p=0.826).
The other explanation hinges on the fact that the survey relied on hikers’ own assessments of whether or not trail water caused them to become ill. It’s plausible that hikers reported getting sick from trail water, but actually got sick for other reasons. The survey attempted to control for uncertainty by offering the previously-mentioned “I don’t know” response. Consequently, hikers giving a yes or no response could have been more certain about the source of their gastrointestinal malady.
Nonetheless, the fact remains that because there was an equal proportion of hikers reporting water sickness in both the WT and NWT groups, and water treatment technology presumably works as designed, it might very well be that the perception of water-caused sickness was actually sickness caused for other reasons, such as poor hand-cleansing techniques and failure to prevent hand-to-mouth fecal contamination between hikers. This was the conclusion reached by Dr. Welch and is currently the conclusion best-supported by this data.
Water baddies such as giardia, cryptosporidium, and e.coli are real and do exist in backcountry water. Studies show that they exist even in pristine areas like the Sierra Nevada and are in higher concentrations in areas where human and pack animal presence is concentrated. However, based on current survey data on long distance hikers, who are transient and not likely to stay and consume great volumes of water in high, multi-use traffic areas, it makes more sense to emphasize good hand hygiene, as Dr. Welch suggested, and de-emphasize the minutia of different water treatment technologies when discussing the prevention of gastrointestinal illness on the AT and PCT.
Want to know about the next post right when it comes out? Subscribe to distancehiking.com by clicking on the subscription button at the bottom of this page! See a mistake with the stats? Contact us.