Full Terms & Conditions of access and use can be found at
Download by: [University of Wisconsin - Madison] Date: 04 November 2015, At: 08:50
Scandinavian Journal of Gastroenterology
ISSN: 0036-5521 (Print) 1502-7708 (Online) Journal homepage: http://www.tandfonline.com/loi/igas20
Stool Form Scale as a Useful Guide to Intestinal
S. J. Lewis & K. W. Heaton
To cite this article: S. J. Lewis & K. W. Heaton (1997) Stool Form Scale as a Useful Guide to
Intestinal Transit Time, Scandinavian Journal of Gastroenterology, 32:9, 920-924
To link to this article: http://dx.doi.org/10.3109/00365529709011203
Published online: 08 Jul 2009.
Submit your article to this journal
Article views: 3327
View related articles
Citing articles: 20 View citing articles
Stool Form Scale as a Useful Guide to Intestinal Transit Time
S. J. LEWIS & K. W. HEATON
University Dept. of Medicine, Bristol Royal Infirmary, Bristol, UK
Lewis SJ, Heaton KW. Stool form as a useful guide to intestinal transit time. Scand J Gastroenterol
Background: Stool form scales are a simple method of assessing intestinal transit rate but are not widely
used in clinical practice or research, possibly because of the lack of evidence that they are responsive to
changes in transit time. We set out to assess the responsiveness of the Bristol stool form scale to change in
transit time. Methods: Sixty-six volunteers had their whole-gut transit time (WGTT) measured with
radiopaque marker pellets and their stools weighed, and they kept a diary of their stool form on a 7-point
scale and of their defecatory frequency. WGTT was then altered with senna and loperamide, and the
measurements were repeated. Results: The base-line WGlT measurements correlated with defecatory
frequency (r = 0.35, P = 0.005) and with stool output (r= -0.41, P = 0.001) but best with stool form
(r= -0.54, P < 0.001). When the volunteers took senna (n= 44), the WGlT decreased, whereas
defecatory frequency, stool form score, and stool output increased (all, P < 0.001). With loperamide
(n = 43) all measurements changed in the opposite direction. Change in WGTT from base line correlated
with change in defecatory frequency (r= 0.41, P < 0.001) and with change in stool output (r= -0.54,
P < 0.001) but best with change in stool form (r= -0.65, P < 0.001). Conclusions: This study has shown
that a stool form scale can be used to monitor change in intestinal function. Such scales have utility in both
clinical practice and research.
Key words: Constipation; diarrhoea; intestinal transit; stool form score
Dr. Stephen Lewis, Dept. of Medicine, University Hospital of Wales, Heath Park, Cardiff CF4 4XW,
Wales, UK (fax: $44 1222-745131)
The rate of passage of intestinal contents is of central
importance in gut physiology and is a major determinant of
bowel symptoms. Fast intestinal transit impairs absorptive
function in both the small and large intestine (1-3), alters
bacterial turnover (4), acidifies the distal colon (9, and leads
to diarrhoea with its distressing symptoms of urgency,
tenesmus, and incontinence. Slow transit markedly alters
bacterial growth and metabolism (4), bile acid metabolism
(6), and oestrogen metabolism (7) and leads to the annoyance
of straining at stool and to many of the symptoms of the
imtable bowel syndrome (IBS) (8). It also increases the risk
of bowel cancer (9) and of gallstones (10). Despite all this,
intestinal transit time is seldom measured either in clinical
practice or in epidemiologic studies. One reason for this is the
cumbersome or unpleasant nature of standard techniques for
measuring transit time, relying, as they do, on collection of
stools or radiation exposure, or both (1 1). However, this
reason may no longer exist.
The recent invention of stool form scales (12, 13) has
provided a simple tool for measuring transit time in non-
hospitalized people. The technique involves no exposure to
radiation, no stool handling, and no discomfort except the
mild unpleasantness of subjects having to inspect their own
faeces and decide which of several descriptions fits best. The
scales have been validated by showing a significant correla-
tion between the scale number of a subject’s stools and the
mean transit time measured previously. With the original
scale this correlation was very close (r = 0.93) (12), but this
must have been partly because all the observations were made
by one trained observer. Her 8-point scale was too compli-
cated for untrained observers, and some of the descriptions
were appropriate only for stools seen on a flat, dry surface and
not for stools seen as they are in real life under water in the
bowl of the water closet. We therefore devised a simpler,
water closet-relevant scale with seven items and with the
descriptions couched in everyday language (Table I). This
‘Bristol Stool Form Scale’ has proved acceptable both to
subjects in epidemiologic surveys and to patients attending
gastrointestinal clinics. Reasonable correlations have been
observed between subjects’ scale scores and their measured
whole-gut transit time (I values around 0.7) (13-15). The
validity of the scale has been confirmed by workers at the
Mayo Clinic (16), and the scale has been recommended for
research by an international working party (17). Its clinical
utility has also been suggested by the finding that symptoms
of straining and urgency are linearly related to scale score
(18). In the clinic or surgery, stool form recordings can be
used to distinguish true diarrhoea and true constipation from
the pseudo-diarrhoea and pseudo-constipation of IBS
Despite all this, stool form scales are not widely used in
clinical practice or research. One reason may be the lack of
evidence that they are responsive to change-in other words,
that, in an individual, a change in transit time is associated
Intestinal Transit Time 921
Table 1. Bristol Stool Form Scale
Table 111. First base-line whole-gut transit time and faecal measure-
ments in volunteers (n = 66) (median, interquartile range, and range)
Type 2 Sausage-shaped but lumpy. Whole-gut transit-time (h) 62.7 47.9, 74.6 20.9, 197.7
Type 3 Like a sausage or snake but with cracks on its surface. Stool output (g/week) 923 646, 1430 104,4218
Type 4 Like a sausage or snake, smooth and soft. Stool form score 3.6 3.2, 4.1 1.4, 5.7
Type 5 Soft blobs with clear-cut edges. Interdefecatory interval (h) 24.0 17.7, 26.7 8.7, 56.0
Separate hard lumps, like nuts.
Fluffy pieces with ragged edges, a mushy stool.
Watery, no solid pieces.
Table 11. Base-line anthropomorphic measurements for volunteers
(n = 66) (median, interquartile range, and range)
Age (years) 40 29, 50 15,62
Height (m) 1.66 1.62, 1.70 1.49, 1.78
Weight (kg) 66.0 60.9, 71.3 47.5, 96.0
Body mass index (kg/m2) 24.0 21.6, 26.4 18.4, 37.7
Waist (cm) 76.0 71.0, 83.0 63.0, 102.0
Hip (cm) 101.0 97.0, 106.3 72.0, 128.0
Waisthip ratio 0.75 0.73, 0.78 0.64, 0.91
with a change in stool form. We have recently used the Bristol
Stool Form Scale extensively in research in which we have
artificially changed the transit time of volunteers. This gave us
the opportunity to assess the responsiveness of stool form to
change in transit time.
MATERIALS AND METHODS
As part of several randomized studies requiring assessment of
colonic function in women (5, 19) 66 healthy omnivorous
volunteers were recruited by advertisements placed in local
hospitals. None had a significant medical history, were obese,
or had taken antibiotics within the past 3 months (Table 11).
The premenopausal women (n = 45) all reported regular
menstrual cycles, and none had lactated within the past 12
months or taken oral contraceptives. In these women all
assessments were done during the early follicular phase of
their menstrual cycle. The postmenopausal women (n = 21,
confirmed by their serum luteinizing hormone (>30 IUA) and
follicular stimulating hormone (>30 IUA) concentration)
were not taking any hormone replacement therapy. At initial
interview the aims of the project and the commitments
required were explained, a medical history was taken, and
height, weight, and waist and hip circumferences were
Volunteers underwent assessment of their base-line colonic
function over 9 days while eating their normal diets. Using
specially designed diaries, the subjects recorded times of
defecation and the ‘form’ or appearance of each stool on a 7-
point scale (1 3,20) ranging from the discrete lumps of slow
transit (type 1) to the non-cohesive (type 6) and liquid stools
(type 7) of rapid transit (Table I).
Whole-gut transit time (WGTT) was measured as a proxy
for colonic transit time by a modification of a published
method (21). Twenty radiopaque marker pellets contained
within a capsule were swallowed (different shapes each day)
on each of four consecutive mornings. The first two stools
passed at least 24 h after ingestion of the last set of markers
were collected, flattened, and roengtenographed. By counting
the number of markers in each stool and then applying the
following formula, the mean WGTT was calculated from the
WG?T (h) = (tlsl + t2~2 .... + ~SSS)/(S~ + ~2 .... + ~g),
where s = the number of markers of a given shape in a stool
sample-that is, 620-and t= the time in hours since
ingestion of this marker pellet to the passing of the stool.
The subscripts 1-8 identify the four different shapes of
marker pellets in the two stool samples. Thus there are up to
four types of pellets in each stool, identified by subscripts 14
in the first and 5-8 in the second stool. Stool output per week
was calculated from the mean weight of the two stools and
Volunteers had up to four sets of base-line measurements
done. These measurements were also repeated after the
volunteers had been randomized to take either senna tablets
(Senokot@, Reckitt & Coleman) or loperamide capsules
(Imodium@, Janssen Pharmaceuticals) for a minimum of 9
days. Both were taken at the maximum tolerated dose.
Volunteers undergoing further cycles of study had a minimum
of a 2-week washout period between agents to obviate any
Data were assessed as parametrically or non-parametrically
distributed, using histograms and Ryan Joiner tests. Changes
were analysed as appropriate by using two-tailed Student’s t
and Wilcoxon rank sum tests. Correlations were calculated
with Spearman’s correlation coefficients, as WGTT was non-
The study was approved by the Research Ethics Committee
of the United Bristol Healthcare Trust.
Replicated base-line measurements (n = 194) within an
individual did not differ significantly from each other. The
coefficients of variation were 17.5% for WGTT, 39.3% for
stool output, 13.6% for stool form, and 17.9% for defecatory
frequency. Thus stool form was the least variable measure-
When the first set of base-line data obtained for each
volunteer was analysed (Table II), WGTT correlated with
922 S. J. Lewis & K. W. Heaton
Table IV. Changes in whole-gut transit time (WGTT) and faecal measurements with senna (n = 44) and loperamide (n = 43) (median, 95%
confidence interval (CI) of the difference)
Change with senna Change with loperamide
Base Active 95% CI P value Base Active 95% CI P value
Interdefecatory interval (h) 24 19 -7.2, -3.2 <0.001 24 31 0.57, 8.3 0.015
Whole-gut transit time (h) 67 41 -26, -14 <0.001 52 72 10.24 <0.001
Stool form score 3.6 4.6 0.77, 1.30 <0.001 3.9 2.8 -1.38, -0.79 <0.001
Stool output (glweek) 784 1482 324,763 <0.001 1143 656 -761, -318 <0.001
defecatory frequency (r = 0.35, P = 0.005) and with stool
output (r= -0.41, P= 0.001) but best with stool form
(r = -0.54, P <0.001).
When volunteers took senna (n=44) or loperamide
(n = 43), all the measured factors of intestinal transit changed
(Table IV). By chance, the base-line measurements of WGTT
(P = 0.045, 95% confidence interval (CI), 0.1, 18.4) was
higher and stool output (P = 0.028, 95% CI, 35.0, 552.4) was
lower before senna than before loperamide. However, in the
23 volunteers who took both senna and loperamide their base-
line measurements of WGTT, stool form, defecatory fre-
quency, and stool output were similar before both agents, and
changes in these factors were all highly significant.
Change in WGTT from the base-line measurements
correlated with change in defecatory frequency (r= 0.41,
P < 0.001) and change in stool output (r = -0.54, P < 0.001)
but best with change in stool form (r = -0.65, P < 0.001)
This study has confirmed again that stool form is a better
predictor of intestinal transit time than defecation frequency.
In fact, in some studies bowel frequency has had no predictive
value at all (13,22,23). We have gone on to show that change
in stool form score correlates with change in transit time. This
implies that stool form scales can be used with confidence
both in the clinical setting and in research. For example, it can
now be confirmed that the widely fluctuating stool form
scores of patients with IBS mean that they have an objective
abnormality in intestinal function (18). A practical implica-
tion is that clinicians can use stool form scales to monitor the
response to treatment of their patients with constipation,
diarrhoea, or IBS.
Given the central importance of transit time in gut physi-
ology and disease, we recommend that recordings of stool
form be incorporated in gastrointestinal research protocols, in
-100 -50 0 50 100
Change in Whole-gut transit-time (h)
Fig. 1. Change in whole-gut transit time and change in stool form of volunteers given senna laxative and
loperamide (r = -0.65).
Intestinal Transit Time 923
which they add virtually nothing to the expense and poten-
tially a good deal to the relevance of the research. As an
example, we used stool forms to show that in the community,
there is a large amount of unsuspected, asymptomatic
constipation (in the sense of slow intestinal transit) and that,
conversely, many people who think they are constipated are
not (24). We have also shown by the use of the Bristol scale
(now also translated by Buddhist monks into Lad&i) that
slow intestinal transit is surprisingly common in a peasant
population eating a very high fibre diet, which might explain
the equally surprising fact that these slim, hardy people are
very prone to cholesterol gallstones (25). In a hospital setting
Britain and the trustees of the United Bristol Health Care
Trust for their generous support of this work. The technical
assistance of Ms Carol Symes is gratefully acknowledged.
1. Read NW, Miles CA, Fisher D. Holgate AM, Kime ND,
Mitchell MA, et al. Transit of a meal through the stomach, small
intestine, colon in and its role in the
pathogenesis of diarrhoea. Gastroenterology 1980;79: 127682.
2. Meihoff WE, Kern F. Bile salt malabsorption in regional ileitis,
ileal resection and mannitol-induced diarrhoea. J Clin Invest
we demonstrated faster intestinal transit (using the Bristol
scale) in a study examining the effects of proton pump
inhibitors on duodenal bacterial counts (26).
The correlation between stool form score and transit time is
very significant but not perfect. There are several possible
reasons for this imperfection. One is observer error; this is
made likely by the very close correlation observed by Davies
et al. (12) in a study in which Davies scored all her volunteers’
stools herself (r= 0.93). Another is that stool form is
determined not so much by overall colonic passage time as
by residence time in one particular segment of the colon, such
as the sigmoid, whose contractions must mould the stool into
its final shape. However, Degen & Phillips (16) found no
correlation between stool form and any particular colonic
segment. A third reason is that there are other unidentified
factors beside transit time. The findings of Davies et al. (12)
make it unlikely such factors are important. If observer error
is the main factor, then there must be scope for improving the
utility of the scale by acclimatizing subjects to it or by
changing its wording so that untrained (and unmotivated)
people are less likely to choose the wrong type number. For
example, type 2 and type 5 are easily confused but could
perhaps be distinguished by reference being made to the
usually easier passage of type-5 stools (1 8).
One possible criticism of this study is that the base-line
measurements before senna reflected a slower intestinal
transit speed than before loperamide and that this could
have biased the findings in our favour. However, the
difference in base-line measurements, a chance finding,
merely reflects the fact that only 23 volunteers were common
to both interventions. In the subgroup of 23 volunteers the
base-line measurements were similar, but equally impressive
changes in measured variables were seen. The correlations
between changes were similar to those in the group as a
In conclusion, this study has conhed the validity of stool
form as a guide to whole-gut transit time and has shown that a
stool form scale can be used to monitor change in intestinal
function. Such scales deserve to be used more widely.
We should like to thank the Kellogg Company of Great
3. Chapman RW. Sillery JK, Graham MM, Saunders DR.
Absorption of starch by healthy ileostomates: effect of transit
time and carbohydrate load. Am J Clin Nutr 1985;41:1244-8.
4. Stephen AM, Wiggins HS, Curnmings JH. Effect of changing
transit time on colonic microbial metabolism in man. Gut
5. Lewis SJ, Heaton KW. Manipulation of intestinal transit rate
alters colonic luminal pH and stool short chain fatty acid
concentration. Gut 1996;39:A40.
6. Marcus SN, Heaton KW. Intestinal transit, deoxycholic acid and
the cholesterol saturation of bile: three inter-related factors. Gut
7. Lewis SJ, Oakey RE, McGarrigle HHG, Heaton KW. Reduction
in serum oestrogens with faster intestinal transit. Gut 1996;39
8. Marcus SN. Heaton KW. Initable bowel-tvDe svmotoms in
spontaneous’ and induced constipation. Gut 1!%7;8:15&9.
9. Cummings JH, Bingham SA, Heaton KW, Eastwood MA.
Faecal weight, colon cancer risk, and dietary intake of nonstarch
polysaccharides (dietary fibre). Gastroenterology 1992; 103:
10. Heaton KW, Emmett PM, Symes CL, Braddon FEM. An
explanation for gallstones in normal-weight women: slow
intestinal transit. Lancet 1993;341:8-10.
1 I. Phillips SF. Measurement of colonic transit. In: Kamm MA,
Lennard-Jones JE, editors. Gastrointestinal transit: pathophysi-
ology and pharmacology. Petersfield (UK): Wrightson; 1991.
12. Davies GJ, Crowder M, Reid B, Dickerson JWT. Bowel
function measurements of individuals with different eating
patterns. Gut 1986;27:1649.
13. O’Donnell LID, Virjee J, Heaton KW. Detection of pseudodiar-
rhoea by simple clinical assessment of intestinal transit rate. Br
Med J 1990;300:439-40.
14. Heaton KW, O’Donnell LJD. An office guide to whole-gut
transit-time: patients recollection of their stool form. J Clin
15. Probert CSJ, Emmett PM, Heaton KW. Some determinants of
whole gut transit-time: a population-based study. Q J Med 1995;
16. Degen LP, Phillips SF. How well does stool form reflect colonic
transit? Gut 1996;39:109-13.
17. Talley NJ, Nyren 0, Drossman DA, Heaton KW, Veldhuyzen
van Zanten SJO, Koch MM, et al. The irritable bowel syndrome:
towards optimal design of controlled treatment trials. Gastro-
enterol Int 1994;6: 189-21 1.
18. Heaton KW, Ghosh S, Braddon FEM. How bad are the
symptoms and bowel dysfunction of patients with irritable
bowel syndrome? A prospective, controlled study with emphasis
on stool form. Gut 1991;32:73-9.
19. Lewis SJ, Bolton C, Heaton KW. Lack of influence of intestinal
transit on oxidative status in premenopausal women. Eur J Clin
20. Probert CSJ, Emmett PM, Heaton KW. Intestinal transit time in
the population calculated from self made observations of
defecation. J Epidemiol Comm Health 1993;47:331-3.
21. Marcus SN, Heaton KW. Effects of a new, concentrated, wheat
924 S. J. Lewis & K. W. Heaton
fibre preparation on intestinal transit, deoxycholic acid metabo-
lism and the composition of bile. Gut 1986;27:893-900.
22. Gorard DA, Gomborone JE, Libby GW, Farthing MJG.
Intestinal transit in anxiety and depression. Gut 1996;39:55 1-5.
23. Krevsky B, Malmud LS, D’Ercole F, Maurer AH, Fisher RS.
Colonic transit scintigraphy. A physiologic approach to the
quantitative measurement of colonic transit in humans. Gastro-
enterology l986;9 1 : 1 102-12.
24. Probert CSJ, Emmett PM, Cripps HA, Heaton KW. Evidence for
Received 4 April 1997
Accepted 6 May 1997
the ambiguity of the term constipation: the role of initable
bowel syndrome. Gut 1994;35:1455-8.
25. Spathis A, Heaton KW, Emmett PM, Norboo T, Hunt L.
Gallstones in a community free of obesity but prone to slow
intestinal transit. Eur J Gastroenterol Hepatol 1997. In press.
26. Lewis SJ, Franco S, Young G, O’Keefe SJD. Altered bowel
function and duodenal bacterial overgrowth in patients treated
with omeprazole. Aliment Pharmacol Ther 1996; 10:557-61.