Human Reproduction vol.13 no.6 pp.1502–1505, 1998
Weight loss in obese infertile women results in
improvement in reproductive outcome for all forms of
A.M.Clark1,4, B.Thornley1, L.Tomlinson2,
1Reproductive Medicine Unit, Department of Obstetrics and
Gynaecology,2Dietetics and3Psychiatry, The Queen Elizabeth
Hospital, University of Adelaide, Woodville, SA 5011 and4Fertility
First, Level 1, Hurstville Community Hospital, 2 Pearl Street,
Hurstville, NSW 2220 Australia
4To whom correspondence should be addressed
Obesity affects ovulation, response to fertility treatment,
pregnancy rates and outcome. In this prospective study, a
weight loss programme was assessed to determine whether
it could help obese infertile women, irrespective of their
infertility diagnosis, to achieve a viable pregnancy,
ideally without further medical intervention. The subjects
underwent a weekly programme aimed at lifestyle
changes in relation to exercise and diet for 6 months;
those that did not complete the 6 months were treated as
a comparison group. Women in the study lost an average
of 10.2 kg/m2, with 60 of the 67 anovulatory subjects
resuming spontaneous ovulation, 52 achieving a pregnancy
(18 spontaneously) and 45 a live birth. The miscarriage
rate was 18%, compared to 75% for the same women
prior to the programme. Psychometric measurements also
improved. None of these changes occurred in the com-
parison group. The cost savings of the programme were
considerable. Prior to the programme, the 67 women had
had treatment costing a total of A$550 000 for two live
births, a cost of A$275 000 per baby. After the programme,
the same women had treatment costing a total of A$210 000
for 45 babies, a cost of A$4600 per baby. Thus weight loss
should be considered as a first option for women who are
infertile and overweight.
Key words: obesity/ovulation/pregnancy/self-esteem/weight
The fertility of obese women compared to normal weight
women is lower in natural cycles and infertility treatment
cycles (Chong et al., 1986; Hamilton-Fairley et al., 1992;
Zaadstra et al., 1993; Crosignani et al., 1994). Higher rates
of miscarriage (Hamilton-Fairley et al., 1992) and congenital
anomalies (Waller et al., 1994) are also reported for this
group of women. In a previous paper (Clark et al., 1995),
we reported that even a small weight loss in anovulatory
obese infertile women, achieved in a group setting over a
6 month period, resulted in an improvement in ovulation,
© European Society for Human Reproduction and Embryology
pregnancy rate and pregnancy outcome, self-esteem and
endocrine parameters. Ninety percent of the anovulatory
women resumed ovulation and 45% had a spontaneous
pregnancy. Of the remaining women who required treatment,
the pregnancy rate per treatment cycle was ?50% per cycle
and the overall miscarriage rate was 25%, compared to 75%
previously for the same women.
The aim of the current study was to apply the same
principles to a larger group of women with mixed indications
for fertility treatment, requiring a wider range of treatment
options. We also wished to do a preliminary cost effectiveness
assessment of the programme.
Materials and methods
Patient inclusion criteria for the study were infertility for ?2
years, body mass index (BMI) ?30 kg/m2, being prepared to take
6 months ‘time out’ from conventional medical treatment for
infertility and being able to attend a 3 h session once a week for
6 months. Subjects with the following attributes were excluded:
presence of a medical condition that would compromise participation
in an exercise programme, presence of an endocrine condition
[other than polycystic ovary syndrome (PCOS)], such as hyperpro-
lactinaemia, thyroid disease or Cushing’s syndrome and a desire
to continue conventional fertility treatment for the duration of the
programme. A total of 120 women who met the criteria were
approached with an information letter and a follow-up telephone
call about the programme. Of these, 87 consented to take part in
the study, which was conducted in four groups, each for 6 months.
The subjects previously reported by Clark et al. (1995) are included
in this report. Eighteen to 30 women started in each group. The
characteristics of the subjects are shown in Table I. The women
were patients at the Reproductive Medicine Unit at The Queen
Elizabeth and Wakefield Hospitals. Of the 87 women, 20 were
unable to complete the 6 month study programme due to work
and other commitments (‘drop-out’ group) and were included for
comparison with those who completed the group programme. There
were no significant differences in age, BMI, length of infertility,
PCOS or ovulation status between the two groups. However, those
that ‘dropped-out’ had had significantly fewer treatment cycles
than those who completed the 6 month programme. Causes of
infertility covered a range of aetiology from anovulation to tubal
disease and male factor infertility. Of the 87 patients, 69 were
anovulatory at the commencement of the study and 53 of the
subjects had some degree of male factor infertility as well.
Treatment, assessment and statistical analysis
These have been described previously (Clark et al., 1995). The fitness
testing and assessment of dietary change detailed previously was not
carried out in this study.
Weight loss improves fertility treatment outcome
Table I. Characteristics of the patients recruited for the study before
participation (values are means ? SD)
(n ? 67)
(n ? 20)
Body mass index (kg/m2)
Duration of infertility (years)
Previous fertility treatments (cycles)
aP ? 0.001.
31.6 ? 4.9
37.4 ? 6.9
5.4 ? 2.5
3.7 ? 1.2
32.8 ? 5.0
35.9 ? 4.1
6.2 ? 2.4
1.0 ? 0.5a
PCOS ? polycystic ovary syndrome.
Table II. Comparison between those who completed and those who did not
complete the study
(n ? 67)
(n ? 20)
Change in body mass index (kg/m2)
Resumed spontaneous ovulation (%)
Total women pregnant (%)c
Total women with live birth (%)
?3.7 ? 1.6
?0.4 ? 1.4a
aP ? 0.001.
bP ? 0.001.
cNine (13%) avoiding treatment.
Women who attended the programme over the 6 months had
significant weight loss (10.2 ? 4.3 kg, range 3.5–15; P ?
0.001). Those who had not conceived 9 months after the end
of the programme maintained this weight loss. In contrast, the
‘drop-out’ group had an insignificant weight loss (1.2 ? 3.6
kg, range ?3.2 to –2.8) (Table II).
In assessing the women’s progress in the Unit prior to
starting the programme, it was noted that they had an average
increase in BMI per year of 1 kg/m2. This is 10 times
the normal expected annual increase of 0.1 kg/m2(Rookus
et al., 1987).
At the beginning of the study, 69 (80%) of the women were
anovulatory as judged by standard endocrine criteria. At the
end of the 6 months, 90% of the previous anovulatory women
in the study group were ovulating spontaneously compared to
none of the ‘drop-out’ group. As previously noted the return
to ovulation occurred after a small weight loss, with all women
who resumed ovulation doing so by the fifth month of the
programme, despite a mean weight loss at that time of 6.5 kg,
which meant that all were still in the obese BMI range of ?30
kg/m2. The anovulatory women who attended ?66% of the
sessions all resumed spontaneous ovulation.
Table III. Pregnancies per cycle after participation in the programme and
occurring simultaneously in non-study patients
(n ? 67)
(n ? 20)
Average pregnancy rate in
the Unit over same period
Pregnancy rate (%)
OI ? ovulation induction, IUI ? intrauterine insemination, IVF ? in-vitro
Values in parentheses are percentages.
Of the 67 women who completed the study, 52 (77.6%)
conceived, 45 (67%) achieving a live birth. Nine (13%)
of the women elected not to proceed with further treatment
cycles due to changes in their social or financial circum-
stances. Eighteen (32.7%) of the 55 pregnancies occurred
spontaneously, the remainder following treatment. The sub-
groups of women in the study who did not do as well in terms
of spontaneous and treatment pregnancies included smokers,
those who attended less than two-thirds of the sessions over
the 6 month programme or whose BMI was still ?40 kg/m2
at the end of the programme. All the women who did not
achieve a spontaneous pregnancy or live birth were in one
or more of these categories. Table III shows the women’s
pregnancy rates on treatment after the programme. Fewer
ovulation induction cycles were carried out following the
programme as it was principally the patients requiring that
mode of treatment who spontaneously conceived. No pregnan-
cies occurred in the ‘drop-out’ group, despite the majority
continuing medical treatment. Of the 33 women who elected
not to participate in the group (principally due to timing during
the working day), but to be followed up, two pregnancies
occurred on subsequent treatment cycles in the following 18
months, despite nine of the women achieving a 5–7 kg weight
loss on their own. There were no spontaneous pregnancies in
this group. Eighty percent of anovulatory women achieved a
pregnancy (40% while in the study, 60% after the study), 63%
of those with tubal factor [all on in-vitro fertilization (IVF)
treatment] and 83% with male factor [50% on IVF or intracyto-
plasmic sperm injection (ICSI)].
Prior to the programme, the 67 women who completed the
study had achieved a total of eight conceptions, of which six
miscarried (75% miscarriage rate). Following the programme,
10 of the 55 pregnancies miscarried [18% miscarriage rate,
(P ? 0.01)].
There was a significant improvement in all the psychological
parameters measured, consistent with a global improvement
in psychological health. In particular, the mean self-esteem
A.M.Clark et al.
score for the study group rose from 19.3 to 21.3 (P ? 0.01).
The mean anxiety score for the study group was reduced from
6.7 down to 5.6 (P ? 0.01). The mean depression score was
reduced from 4.1 down to 2.2 (P ? 0.001).
with all four groups maintaining informal exercise sessions
and meetings throughout and after the course.
The total cost of running the 6 month programme (hours
worked by each individual plus administration) was 8828
Australian dollars (A$). In comparison, the cost of one IVF
cycle averaged A$4150 and one gonadotrophin ovulation
induction cycle, calculated using the average number of
ampoules used by these women, was A$1050. Therefore, the
saving of two IVF cycles or eight ovulation induction cycles
would have funded the programme in this Unit.
Prior to the programme, the 67 women who completed the
study had had treatment totalling A$550 000 for two live
births, a cost of A$275 000 per baby. After participation in
the programme, the same 67 women had treatment costing
A$210 000 for 45 live births, at a cost of A$4600 per baby.
This study, which is an extension of a study previously reported
involving women requiring ovulation induction, demonstrates
that a group approach to the combined problem of obesity
and infertility is associated with a marked improvement in
pregnancy and ovulation rates and a reduction in the need for
the use of high technology treatment. Obese infertile women,
irrespective of their infertility diagnosis, appear to benefit. The
outcomes in terms of pregnancy and ovulation rates were
greater than could be expected based on the patients’ past
histories, and these changes, in combination with the signific-
antly lowered miscarriage rate, indicate the programme is
clearly cost-effective compared to starting conventional med-
ical treatment for obese infertile women when they first present.
There was a marked disparity in outcome between those
women who failed to complete the programme and the study
group who finished the 6 months despite having access to the
same information during the 2–3 months they attended the
programme. In addition, failure to attend more than two thirds
of the sessions was associated with a less positive reproductive
The possible explanations for these results are still uncon-
firmed but recent publications on the impact of insulin indicate
by weight loss or administration of an oral hypoglycaemic
results in spontaneous ovulation (Velazquez et al., 1994). In
our preliminary study, we showed that women who reduced
their weight had lower insulin concentrations. Others (Guzick
et al., 1994; Helmen et al., 1996) have shown the same effect
of weight loss on insulin concentrations and pregnancy rates.
It is unclear whether reduction in insulin is the sole contributor
to the change in reproductive outcome we found in this study.
Women who were unable to attend the programme but
achieved a similar weight loss independent of the group did
not have the same pregnancy results, suggesting that some
other component of the group process of psychological changes
also affects the results. Others have reported the positive
benefits of improving psychological parameters in relation to
reproductive outcome (Domar et al., 1990; Thiering et al.,
1993), in contrast to Harlow et al. (1996), who assessed a
group of women undergoing IVF treatment and found no
difference in pregnancy outcome between those who registered
higher concentrations of stress hormones compared to the rest
of the IVF population. Stunkard et al. (1980) indicated that
attempts at maintaining weight loss were much more successful
when approached in a group situation than when the same
information was given on a one to one basis. In addition,
behavioural therapy was more successful than the use of drugs
to lose weight.
This study was initiated by our concern for the long term
physical and psychological health of obese infertile patients.
We observed, as have others, that their pregnancy rates were
reduced (Chong et al., 1986; Zaadstra et al., 1993; Crosignani
et al., 1994), their need for higher doses of medication was
increased (McClure et al., 1992) and their increase in weight,
while patients of the Unit, was 10 times the average annual
increase (Rookus et al., 1987; Clark et al., 1995). When
starting the group, we believed that even if the women left
the Unit without getting pregnant, if we had assisted in
improving their long-term physical and psychological health,
the programme would have been a success. The changes in
reproductive outcome have been so striking that this study has
been the basis for a randomized controlled trial of the effects
of weight loss on fertility and treatment outcomes in a group
situation. In the interim, these results continue to support the
view that all who treat infertility should consider weight loss
to be a prerequisite for obese women prior to any assisted
We thank Helen Holmes for typing this report. The Reproductive
Medicine Unit and the University of Adelaide are acknowledged for
Chong, A.P., Rafael, R.W. and Forte, C.C. (1986) Influence of weight in the
induction of ovulation with human menopausal gonadotropin and human
chorionic gonadotropin. Fertil. Steril., 46, 599–603.
Clark, A.M., Ledger, W., Galletly, C. et al. (1995) Weight loss results in
significant improvement in pregnancy and ovulation rates in anovulatory
obese women. Hum. Reprod., 10, 2705–2712.
Crosignani, P.G., Ragni, G., Parazzini, F. et al. (1994) Anthropometric
indicators and response to gonadotropin for ovulation induction. Hum.
Reprod., 9, 420–423.
Domar, A.D., Seibel, M.M. and Benson, H. (1990) The Mind/Body Program
for Infertility: a new behavioural treatment approach for women with
infertility. Fertil. Steril., 53, 246–249.
Guzick, D.S., Wing, R., Smith, D. et al. (1994) Endocrine consequences of
weight loss in obese, hyperandrogenic, anovulatory women. Fertil. Steril.,
Hamilton-Fairley, D., Kiddy, D., Watson, H. et al. (1992) Association of
moderate obesity with a poor pregnancy outcome in women with polycystic
ovary syndrome treated with low dose gonadotrophin. Br. J. Obstet.
Gynaecol., 99, 128–131.
Weight loss improves fertility treatment outcome
Harlow, C.R., Fahy, U.M., Talbot, W.M. et al. (1996) Stress and stress-related
hormones during in-vitro fertilization treatment. Hum. Reprod., 11, 274–279.
Helman, M., Runnebaum, B. and Gerhard I. (1996) Effects of weight loss on
McClure, N., McQuinn, B., McDonald, J. et al. (1992) Body weight, body
mass index, and age: predictors of menotropin dose and cycle outcome in
polycystic ovarian syndrome? Fertil. Steril., 58, 622–624.
Rookus, M.A., Burema, J., Hol Van, M.A. et al. (1987) The development of
body mass index in young adults. II. Interrelationships of level, change and
fluctuation, a four-year longitudinal study. Hum. Biol., 59, 617–630.
Stunkard, A.J., Craighead, L.W. and O’Brien, R. (1980) Controlled trial of
behaviour therapy, pharmacotherapy and their combination in the treatment
of obesity. Lancet, 2, 1045–1047.
Thiering, P., Beaurepaire, J., Jones, M. et al. (1993) Mood state as a predictor
of treatment outcome after in vitro fertilisation/embryo transfer technology
(IVF/ET). J. Psychosomat. Res., 37, 481–491.
Velazquez, E.M., Mendoza, S., Hamer, T. et al. (1994) Metformin therapy in
polycystic ovary syndrome reduces hyperinsulinemia, insulin resistance,
hyperandrogenemia and systolic blood pressure, while facilitating normal
menses and pregnancy. Metabolism, 43, 647–654.
Waller, D.K., Mills, J.L., Simpson, J.L. et al. (1994) Are obese women at
higher risk for producing malformed offspring? Am. J. Obstet. Gynecol.,
Zaadstra, B.M., Seidell, J.C., Van Noord, P.A.H. et al. (1993) Fat and female
fecundity: prospective study of effect of body fat distribution on conception
rates. Br. Med. J., 306, 484–487.
Received on August 26, 1997; accepted on February 23, 1998