The eating-disorders chapter in Gut and Psychology Syndrome opens with several strong statistics. To use them well, you need to keep three questions separate: how common a condition is, how mortality is measured and what a study calls relapse. Those numbers have different denominators and cannot stand in for one another.[1]
The Figures in the GAPS Chapter
The expanded text represented by the second German edition in 2015 makes the following statements. These are the chapter’s historical descriptions, not a new survey of the population in 2026.[1]
| Chapter Statement | Context Needed |
|---|---|
| Eating disorders cause more deaths than any other mental illness. | The passage does not give a numeric death rate or a study comparing every current psychiatric diagnosis. |
| About 90 percent of affected people are girls ages 12 to 25. | This is the chapter’s general demographic statement. It is not a rule for recognizing who can be ill. |
| About 1 percent of the population is affected. | The passage does not specify a country, survey year, diagnostic definition or lifetime versus current prevalence window. |
| Relapse is estimated at 50 percent or more. | The passage does not define one diagnosis, treatment, follow-up interval or operational definition of relapse for this estimate. |
The chapter describes much higher numbers in wealthy industrialized countries than elsewhere. That is its historical geographic comparison; it does not give country-specific rates or a measured size for that difference. The chapter also says many cases remain undiagnosed or unrecorded because of shame, secrecy and fear of rejection. It does not assign a separate percentage to those unrecorded cases. Its bibliography lists research, but the opening passage does not connect each headline figure to a separately defined cohort.
Mortality Ratios Are Not Percentages
Arcelus and colleagues’ 2011 meta-analysis examined 36 studies published between 1966 and September 2010. It reported standardized mortality ratios of 5.86 for anorexia nervosa, 1.93 for bulimia nervosa and 1.92 for eating disorder not otherwise specified, or EDNOS.[2]
A standardized mortality ratio compares observed deaths with the number expected in a reference population. A ratio of 5.86 does not mean that 5.86 percent of everyone with anorexia died. The diagnosis, study population and follow-up still matter.
These results support taking eating disorders seriously. They do not establish a ranking against every other mental illness in every year, and they do not predict the outcome of a particular person receiving care.
Prevalence Needs a Diagnosis and Time Window
One useful example is Udo and Grilo’s 2018 analysis of the 2012–2013 U.S. National Epidemiologic Survey on Alcohol and Related Conditions. It used DSM-5 diagnostic definitions and reported separate lifetime and 12-month estimates for different eating disorders.[3]
| Diagnosis | Lifetime Estimate | 12-Month Estimate |
|---|---|---|
| Anorexia nervosa | 0.80% | 0.05% |
| Bulimia nervosa | 0.28% | 0.14% |
| Binge-eating disorder | 0.85% | 0.44% |
Those are estimates from that adult survey and its definitions, not today’s rate for every age group or every feeding disorder. They show why “1 percent of the population” needs more explanation before it can answer a specific question.
Who Can Be Missed by a Demographic Headline?
The book’s own passage says men as well as women experience eating disorders. An age or sex headline should not become a reason to dismiss someone who falls outside it. NIMH describes eating disorders across ages, sexes and body weights, including people who appear healthy.[1][4]
If you are concerned about someone’s eating, the relevant question is what is happening to that person. Matching a demographic description does not diagnose an eating disorder, and failing to match it does not rule one out.
Relapse Rates Depend on the Definition
The GAPS chapter’s “50 percent or more” estimate is not presented as one exact rate for every diagnosis. Before comparing it with another figure, look for the original definition: did relapse mean a weight change, a return of particular behaviors, renewed diagnostic criteria or another treatment episode? Also check when follow-up began and how long it lasted.
The chapter uses the estimate to introduce its argument for examining nutrition and digestive health alongside psychological factors. The percentage itself does not test whether GAPS prevents relapse or establish that food composition caused a particular recurrence.[1]
Overlapping Diagnoses Need Their Own Context
Campbell-McBride discusses movement between anorexia, bulimia, binge eating and compulsive overeating, and overlap with other mental-health conditions. The chapter lists ADHD, obsessive-compulsive disorder, bipolar disorder, panic, anxiety, substance misuse and schizophrenia. It does not give a percentage for each pairing.[1]
A changed diagnosis over time, a co-occurring diagnosis and a symptom shared by two conditions are different findings. Keeping the actual history in order is more useful than combining them into one statistic.
Questions to Ask When You Read a Statistic
- Which condition was measured, and how was it defined?
- Was the sample drawn from a clinic, hospital or population survey?
- Which ages, country and study years were included?
- Does the number describe people, deaths, person-years or a ratio?
- Was the time window current, 12-month or lifetime?
- For relapse, what counted as recovery first and what counted as a recurrence?
You do not need to become a statistician to use those questions. They help you see whether two numbers are describing the same thing.
Your Next Step
Use the eating-disorders overview for the book’s wider argument and the Hanna case history for the difference between an individual story and population evidence. My Care Team Log is listed in the WFLI Shop. A notebook can hold the study title, your question and the explanation you want to discuss.
References
- Campbell-McBride N. Gut and Psychology Syndrome, expanded text represented by the second German edition (2015), “Essstörungen” (“Eating Disorders”). English paraphrases checked against that text; no English page numbers are implied. Opening mortality, demographic, prevalence, overlap and relapse passages.
- Arcelus J et al. Mortality rates in patients with anorexia nervosa and other eating disorders: a meta-analysis of 36 studies. Archives of General Psychiatry. 2011;68:724–731.
- Udo T, Grilo CM. Prevalence and Correlates of DSM-5 Eating Disorders in a Nationally Representative Sample of United States Adults. Biological Psychiatry. 2018;84:345–354. Estimates from 2012–2013 survey data.
- NIMH. Eating Disorders: What You Need to Know. Affected populations and body weights.
Leave a comment