Prescription Use of Paracetamol and Risk for Ovarian Cancer in Denmark
JNCI J Natl Cancer Inst (
Prescription Use of Paracetamol and Risk for Ovarian Cancer in Denmark
BRief C OmmUni Cati On 0
Louise Baandrup 0
Søren Friis 0
Christian Dehlendorff 0
Klaus K. Andersen 0
Jørgen H. Olsen 0
Susanne K. Kjaer ) 0
0 Center , Strandboulevarden 49, Copenhagen DK-2100 , Denmark (
It has been suggested that paracetamol reduces the risk for ovarian cancer.We examined the association between prescription use of paracetamol and ovarian cancer risk in a nationwide case-control study nested within the Danish female population. Case patients (n = 3471) were all women with a first diagnosis of epithelial ovarian cancer during the period from 2000 to 2009. Population control subjects (n = 50 576) were selected by risk set sampling. Data were derived from prescription and other nationwide registries. Conditional logistic regression was used to estimate odds ratios (ORs) and two-sided 95% confidence intervals (CIs) for ovarian cancer associated with use of paracetamol or nonaspirin nonsteroidal anti-inflammatory drugs (NSAIDs). All statistical tests were two-sided. Use of paracetamol was associated with a reduced odds ratio for ovarian cancer (OR = 0.82; 95% CI = 0.74 to 0.92; P < .001) compared with nonuse, and the odds ratio decreased further with long-term (≥10 years), high-intensity paracetamol use (OR = 0.45; 95% CI = 0.24 to 0.86; P = .02). Use of nonaspirin NSAIDs was not associated with ovarian cancer risk.
In 1998, Cramer et al. reported an inverse
association between use of paracetamol and
ovarian cancer (
). The largest reduction
was found among women who had used
paracetamol daily for more than 10 years.
Subsequent observational studies of the
association between paracetamol and
ovarian cancer have yielded inconclusive results
); however, most studies have been
small, based on self-report, or have had
The aim of this study was to evaluate
the association between prescription use
of paracetamol and risk for ovarian cancer.
For comparison, we included analyses of
prescription use of nonaspirin nonsteroidal
anti-inflammatory drugs (NSAIDs), which
have similar indications for pain relief as
We conducted a case–control study
nested in the entire Danish female
population (2.7 million) using data from seven
nationwide registries holding information
on cancer diagnoses (
), prescription use
), hospital contacts (
), and education (
between the registries was performed by
means of the unique civil registration
number assigned to all Danish citizens (
Case patients were all women aged 30 to
84 years recorded in the Danish Cancer
) with a first diagnosis of
histologically verified ovarian cancer during the
period from 2000 to 2009 and with no prior
history of cancer (except nonmelanoma
skin cancer). For each case, we randomly
selected 15 female age-matched
population control subjects (n = 50 576) by risk set
sampling among women with no history of
cancer or bilateral oophorectomy before
the index date (ie, date of case diagnosis and
corresponding date for control subjects).
From the Danish Prescription Registry,
which holds detailed information on
prescription drugs dispensed at Danish
pharmacies since 1995 (
), we identified all
prescriptions for paracetamol and
nonaspirin NSAIDs redeemed by case patients and
control subjects between 1995 and 2009.
We disregarded use within 1 year of the
index date to reduce possible reverse
). Ever use of paracetamol was
defined as 2 or more prescriptions on
separate dates, whereas less than 2 prescriptions
(nonuse) served as reference. Duration
of use was defined as the period between
the first and last prescription plus 60 days
and categorized as less than 5 years, 5 to
10 years, and more than 10 years. Intensity
of use was defined as the cumulative
number of defined daily doses (23) divided by
the duration of use in days and classified
into approximate tertiles of low, medium,
and high intensity.
Conditional logistic regression was
used to estimate age- and
multivariableadjusted odds ratios (ORs) and two-sided
95% confidence intervals (CIs) for ovarian
cancer associated with use of paracetamol
or nonaspirin NSAIDs. All statistical tests
were two-sided (significance level = .05).
Analyses were performed with the
statistical software R versions 2.15.3 (
Further details of the methods,
including codes used in the analysis, are
presented in the Supplementary Methods and
Supplementary Table 1 (available online).
We identified 3471 epithelial ovarian
cancer case patients (n = 2262 with serous
cancer; n = 577 with endometrioid cancer;
n = 411 with mucinous cancer; and n = 221
with clear-cell cancer). The characteristics
of case patients and age-matched control
subjects are presented in Table 1. The
prevalence of paracetamol prescription use
was 13.2% among case patients and 15.1%
among control subjects. Among all 8084
paracetamol users, the median number
of prescriptions was eight (range = 2–551
Ever use of paracetamol was
associated with a statistically significant reduced
odds ratio for epithelial ovarian cancer
(OR = 0.82; 95% CI = 0.74 to 0.92; P <
.001) and serous ovarian cancer (OR = 0.82;
95% CI = 0.72 to 0.94; P = .003) compared
with nonuse (Table 2). Similar risk
reductions were observed for the remaining
histologic subtypes of epithelial ovarian
cancer, albeit with lower statistical
precision (Supplementary Table 2, available
online). The odds ratios for overall
epithelial and serous ovarian cancer decreased
with increasing duration and estimated
daily dose of paracetamol (Table 2). For
long-term (≥10 years), high-intensity
paracetamol use, the odds ratios were 0.45
(95% CI = 0.24 to 0.86; P = .02) for
epithelial ovarian cancer and 0.37 (95% CI = 0.16
to 0.84; P = .02) for serous ovarian
cancer (Table 2). Similar risk estimates were
obtained when the analyses were restricted
to recent use of paracetamol (data not
The association between ever use of
nonaspirin NSAIDs and epithelial ovarian
cancer was close to unity. Results did not
differ by histologic subtype of epithelial
ovarian cancer or by duration and intensity
of nonaspirin NSAID use (Supplementary
Table 3, available online).
Similar results to those of the main
analyses were found after additional adjustment
for family history of ovarian and breast
cancer (information only available for women
born after 1953) and in a sensitivity
analysis applying a new-user design (
details, see the Supplementary Methods,
available online). Finally, although use of
2 of 5 Brief Communication | JNCI
paracetamol and nonaspirin NSAIDs were
correlated (correlation coefficient = .26; P <
.001), nonaspirin NSAID use did not
influence the associations between paracetamol
use and ovarian cancer risk among
longterm, high-intensity paracetamol users
(P = .85).
The inverse association between
paracetamol use and ovarian cancer risk
observed in our study is consistent with a
previous meta-analysis in which an overall
inverse association was reported, with an
indication of a dose–response relationship
between paracetamol use and ovarian
cancer risk (
). Overall, most studies of
paracetamol use and ovarian cancer risk have
reported slightly reduced relative risk
), and two studies
found strong inverse associations similar to
that in our study (
We included analyses of nonaspirin
NSAIDs because paracetamol and
nonaspirin NSAIDs have mutual indications. In
line with our recent meta-analysis (
observed no association between
nonaspirin NSAID use and ovarian cancer. Because
of the heterogeneity within the therapeutic
group of nonaspirin NSAID agents, our
results do not rule out risk variation
according to specific types of nonaspirin NSAIDs.
However, the fact that we observed
substantially reduced risk estimates for
paracetamol but not for nonaspirin NSAIDs
suggests that our case patient group was
not systematically biased with respect to
use of analgesics.
Some biological explanations for a
potential antineoplastic effect of
paracetamol against ovarian cancer have been
offered. Paracetamol has been suggested to
possess antigonadotropic properties due to
chemical similarities with the sex steroids
estradiol and progesterone (
Further, the metabolism of paracetamol
requires glutathione, which is also required
for both the release and receptor binding of
follicle-stimulating hormone. Glutathione
depletion from paracetamol metabolism
may thus reduce the effective
concentration of follicle-stimulating hormone (
Finally, it has also been suggested that
paracetamol metabolites may inhibit the
) ) )
4 7 9
)t .12 .1 .
,fto can tsed I)C
n n ju % ren to to to to to ren to to to
cee ira -ad (95 fee .76 .71 .78 .51 .96 fee .80 .69 .62
r a e R (r (0 (0 (0 (0 (0 (r (0 (0 (0
, v g
r o A O 0 5 1 2 1 8 0 9 6 8
c E l N
acn rton ,tsc 1952 5627 8075 7542 #― **― 1592 2731 7451 7241
se ,ts 2 9 2 7
C ie 3
).t001 .).)0942802 .).)9028709 .).)0190906 .).)9080706 .).)0118901 ).t001 .).)1226909 .).)0164804 .).)9086705 .11 )(.)(rftrfrtrneee001eeen .).(...)(tto96811078067o1150 ..)(...)(tto86021078065o1160 .).(...)(tto45101017052o0970 .33 )(.)(rftrfrtrneee001eeen .).(...)(tto24137057042o1350 ..)(...)(tto94410008047o1370 ..)(...)(tto52808054024o0860 .52 illiiiiI . trrfscassccgoegenononodeen=C li/(),rttrrrfaccvavsvsnopeeoeneeeueom iilirttrrrsccvaysasobnueohnoupedeom /().rrvvsneeeeg i.ttxadedneeh iii .trrfrtts–yacssxanop31eeoebhnedede liliiil/rtrrasaavscccassa6222eouehnoenehew liili/trrrasvsacaccasasehp2262e1ouoenneew iill .rtrtascsssyaeppondd06up iiiiltftrvysaasyascasdedbheddonounedndu iili::,.tttysaasysdendneeod81002o90DDDm
lrtsceaaeauom..tsauoop llltrrcsaeeveaaaoupm .s410onue‡10N .rvs54eue§08E .tcs92eenue80Rǁ .rrs61Foeu90e¶m il().rsays#pedoe70DDDD― )(.**90itrrryaeuaonpe―D lifttrrrsscaeey5aaeaa††<uopuonom .s42310onue‡01N ii-.ttsys69ue‡‡Lonen90w iii-.ttsys49‡une‡eedun80mM iii-.ttssy97ue‡‡ghnen70H Ptrned lifttrrrs–scaey05e1aaeaa††uopouonm .s429511230onue‡001N ii-.ttsys91359ue‡‡Lonen090w iii-.ttsys79251‡une‡eedun090mM iii-.ttsys81243ue‡‡ghnen470H Ptredn lifttrrrsscaeye10aaeaa††>uopouonm .s422951301onue‡001N ii-.ttsys22151ue‡‡Lonen670w iii-.ttsys52491‡une‡eedun380mM iii-.ttsys02951ue‡‡ghnen740H Ptrend iliiii-rtrfttrtsavcascssascaeneeebedononddeed95dnoddodnonOww%iiiiii. ; trtfrtasacsaddooondonn=ROm iiji,)((,,,),rttrfrtfrtyacayaassasspgn0h123bgeoeduededdooeu≥Owm lliiilili),/(,(/)(/,)trtrrtrttrrrzsasvvvasavvcveebedueeeneeeonebueeeneeenm iilliiiii/,)(-rttfrrrrrtrasaasavvassayadoennonnnpnonndeneeedoepnudomm lliili.trrrtrfscvsaacaaassasnopoeepeopnede2donue<Nwm iili.trrrttrftrcssaaacaaasvsasassnoponeppeeoepenedd2eueed≥Ewm iili trrrrtfrftscs–yascaaasaascspepno13eep2denedeenueeoebo≥Rwm lliilitrrrrtfrtrrscsvacaaasaasspepnooebuep2denedFoeueep2o≥<wmm ililitfrtrrvasacassasssaasaysouunbnoeoheedeodeppe1734eDDDDw iliitrfrrrtrtfcsvasaaassssasaysaaouunnobeoeedpeeeuono734ehuDw iiilirtttftrtfrrtfsasacaaasaasoephedbeeenhdedneeeouuononepDwwm liiliIttftrfrttfcasavsacaaassyuuheeunbednedeeeounneoopDDDwmmm iiil.:,:..;trtrtsacss0o30015egeneneep10o90doe009nudDDDDDDmm li..:.rtrtsac901onge047e90peenDDD
* † ‡
P O D D D * † ‡ § ǁ ¶ # * † ‡
an iss se
rfo itro ,)u
e e n
o om now
ilay end kun
ld ),r l,
ca ve an
ityp /ren itao
th (ve ,v
iicshh liitrtye i,hgeh
,sew i),rn icsa
o e (b
d ve n
liya /rn itao
d ve c
d (e ud
ifne se ,)e
e n r
d o ve
= rm en
D oh /r
l; s (e
a au a
rv p m
e o h
n en tsa
cytokine macrophage migration factor,
which is an important regulator of the
inflammatory response (
Our study had several strengths,
including its size, nested case–control design
within the entire female Danish
population, and virtually complete and accurate
cancer ascertainment from the Cancer
Registry. Additional strengths include the
setting with free access to health services
and continuously updated, accurate data on
prescriptions, medical conditions,
reproductive factors, and education, thus
covering most ovarian cancer risk factors. Use of
nationwide registries eliminated recall bias
and minimized selection bias. Cancer
diagnoses were restricted to histologically
verified cases, further enhancing validity (
If paracetamol does possess antineoplastic
effects against ovarian cancer, careful risk–
benefit considerations are needed before
any chemopreventive measures can be
introduced because of the potential adverse
effects of long-term paracetamol use.
The main limitation of our study was
the lack of information of over-the-counter
use of paracetamol, which may have
introduced residual confounding and
misclassification bias. During the study period,
approximately 40% to 45% of purchased
paracetamol and 80% to 85% of nonaspirin
NSAIDs were prescribed (
). If women
receiving paracetamol by prescription
differed from other women with respect to
ovarian cancer risk factors, confounding
might result. However, we were able to
adjust for most established ovarian cancer
risk factors. Moreover, additional analyses
comparing the prescription users of
paracetamol with women with no recorded use
of analgesic drugs yielded risk estimates for
ovarian cancer similar to those of the main
analysis. Finally, the prevalence of
overthe-counter use of paracetamol was likely
lower among chronic users because 50% of
the cost of these agents is reimbursed when
they are prescribed by a physician. Still,
some residual confounding might have
resulted from the lack of information on
Another limitation was potential
noncompliance. However, additional analyses
restricted to continuous use (≥2
prescriptions per year) of paracetamol yielded
results similar to the main analyses,
indicating that noncompliance among
long-term paracetamol users was not a
4 of 5 Brief Communication | JNCI
major issue in this study. Furthermore,
we reduced potential misclassification
due to use of paracetamol and nonaspirin
NSAIDs for symptoms of a
yet-undiagnosed ovarian cancer by applying a 1-year
lag period and evaluated the influence of
use before the start of the Prescription
Registry by applying a new-user design.
Finally, we only had information on use of
oral contraceptives from 1995. However,
similar risk estimates were found among
users and nonusers of oral contraceptives
in additional subanalyses, indicating that
residual confounding by use of oral
contraceptives was not a major limitation in
In conclusion, we found an inverse
association between prescription paracetamol
use and ovarian cancer risk, which was
strongest for long-term, high-intensity use.
Additional large studies with long
followup and comprehensive information on
paracetamol use and ovarian cancer risk factors
are needed to confirm our results.
1. Cramer DW , Harlow BL , Titus-Ernstoff L , Bohlke K , Welch WR , Greenberg ER . Overthe-counter analgesics and risk of ovarian cancer . Lancet . 1998 ; 351 ( 9096 ): 104 - 107 .
2. Fairfield KM , Hunter DJ , Fuchs CS , Colditz GA , Hankinson SE . Aspirin, other NSAIDs, and ovarian cancer risk (United States) . Cancer Causes Control . 2002 ; 13 ( 6 ): 535 - 542 .
3. Pinheiro SP , Tworoger SS , Cramer DW , Rosner BA , Hankinson SE . Use of nonsteroidal antiinflammatory agents and incidence of ovarian cancer in 2 large prospective cohorts . Am J Epidemiol . 2009 ; 169 ( 11 ): 1378 - 1387 .
4. Friis S , Nielsen GL , Mellemkjaer L , et al. Cancer risk in persons receiving prescriptions for paracetamol: a Danish cohort study . Int J Cancer . 2002 ; 97 ( 1 ): 96 - 101 .
5. Ammundsen HB , Faber MT , Jensen A , et al. Use of analgesic drugs and risk of ovarian cancer: results from a Danish casecontrol study . Acta Obstet Gynecol Scand . 2012 ; 91 ( 9 ): 1094 - 1102 .
6. Hannibal CG , Rossing MA , Wicklund KG , Cushing-Haugen KL . Analgesic drug use and risk of epithelial ovarian cancer . Am J Epidemiol . 2008 ; 167 ( 12 ): 1430 - 1437 .
7. Meier CR , Schmitz S , Jick H . Association between acetaminophen or nonsteroidal antiinflammatory drugs and risk of developing ovarian, breast, or colon cancer . Pharmacotherapy . 2002 ; 22 ( 3 ): 303 - 309 .
8. Rosenberg L , Palmer JR , Rao RS , et al. A case-control study of analgesic use and ovarian cancer . Cancer Epidemiol Biomarkers Prev . 2000 ; 9 ( 9 ): 933 - 937 .
9. Moysich KB , Mettlin C , Piver MS , Natarajan N , Menezes RJ , Swede H . Regular use of analgesic drugs and ovarian cancer risk . Cancer Epidemiol Biomarkers Prev . 2001 ; 10 ( 8 ): 903 - 906 .
10. Schildkraut JM , Moorman PG , Halabi S , Calingaert B , Marks JR , Berchuck A . Analgesic drug use and risk of ovarian cancer . Epidemiology . 2006 ; 17 ( 1 ): 104 - 107 .
11. Wu AH , Pearce CL , Tseng CC , Templeman C , Pike MC . Markers of inflammation and risk of ovarian cancer in Los Angeles County . Int J Cancer . 2009 ; 124 ( 6 ): 1409 - 1415 .
12. Lo-Ciganic WH , Zgibor JC , Bunker CH , Moysich KB , Edwards RP , Ness RB . Aspirin, nonaspirin nonsteroidal anti-inflammatory drugs, or acetaminophen and risk of ovarian cancer . Epidemiology . 2012 ; 23 ( 2 ): 311 - 319 .
13. Lacey JV Jr, Sherman ME , Hartge P , Schatzkin A , Schairer C . Medication use and risk of ovarian carcinoma: a prospective study . Int J Cancer . 2004 ; 108 ( 2 ): 281 - 286 .
14. Setiawan VW , Matsuno RK , Lurie G , et al. Use of nonsteroidal anti-inflammatory drugs and risk of ovarian and endometrial cancer: the multiethnic cohort . Cancer Epidemiol Biomarkers Prev . 2012 ; 21 ( 9 ): 1441 - 1449 .
15. Gjerstorff ML . The Danish Cancer Registry . Scand J Public Health . 2011 ; 39 ( suppl 7 ): 42 - 45 .
16. Kildemoes HW , Sorensen HT , Hallas J. The Danish National Prescription Registry. Scand J Public Health . 2011 ; 39 ( Suppl 7 ): 38 - 41 .
17. Lynge E , Sandegaard JL , Rebolj M. The Danish National Patient Register. Scand J Public Health . 2011 ; 39 ( Suppl 7 ): 30 - 33 .
18. Blenstrup LT , Knudsen LB . Danish registers on aspects of reproduction . Scand J Public Health . 2011 ; 39 ( Suppl 7 ): 79 - 82 .
19. Jensen VM , Rasmussen AW. Danish Education Registers. Scand J Public Health . 2011 ; 39 ( Suppl 7 ): 91 - 94 .
20. Pedersen CB. The Danish Civil Registration System. Scand J Public Health . 2011 ; 39 ( Suppl 7 ): 22 - 25 .
21. Rothman K , Greenland S , Lash TL . Casecontrol studies . In: Rothman K , Greenland S , Lash TL , ed. Modern Epidemiology . 3rd ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2008 : 111 - 127 .
22. Jorgensen TL , Herrstedt J , Friis S , Hallas J . Polypharmacy and drug use in elderly Danish cancer patients during 1996 to 2006 . J Geriatr Oncol. 2012 ; 3 ( 1 ): 33 - 40 .
23. World Health Organization Collaborating Centre for Drug Statistics Methodology. ATC Classification Index with DDDs . http://www. whocc.no/atc_ddd_index/. Accessed July 17 , 2013 .
24. R Development Core Team. R: A language and environment for statistical computing . Vienna, Austria: R Foundation for Statistical Computing. http://www.R-project.org/. Accessed July 9 , 2013 .
25. Ray WA . Evaluating medication effects outside of clinical trials: new-user designs . Am J Epidemiol . 2003 ; 158 ( 9 ): 915 - 920 .
26. Bonovas S , Filioussi K , Sitaras NM . Paracetamol use and risk of ovarian cancer: a meta-analysis . Br J Clin Pharmacol . 2006 ; 62 ( 1 ): 113 - 121 .
27. Baandrup L , Faber MT , Christensen J , et al. Nonsteroidal anti-inflammatory drugs and risk of ovarian cancer: systematic review and meta-analysis of observational studies . Acta Obstet Gynecol Scand . 2013 ; 92 ( 3 ): 245 - 255 .
28. Cramer DW , Liberman RF , Hornstein MD , et al. Basal hormone levels in women who use acetaminophen for menstrual pain . Fertil Steril . 1998 ; 70 ( 2 ): 371 - 373 .
29. Gates MA , Tworoger SS , Eliassen AH , Missmer SA , Hankinson SE . Analgesic use and sex steroid hormone concentrations in postmenopausal women . Cancer Epidemiol Biomarkers Prev . 2010 ; 19 ( 4 ): 1033 - 41
30. Senter PD , Al-Abed Y , Metz CN , et al. Inhibition of macrophage migration inhibitory factor (MIF) tautomerase and biological activities by acetaminophen metabolites . Proc Natl Acad Sci U S A . 2002 ; 99 ( 1 ): 144 - 149 .
31. Statens Serum Institut. Medstat. http://www. medstat.dk/. Accessed July 9 , 2013 .