Table 2.

Characteristics of studies on incidence of cancers other than cervical, and on cervico-vaginal cancer mortality after treatment of cervical intraepithelial neoplasia (CIN)

Author, year	Country	Study design	N treated	Treatment method	Degree of treated CINa	Follow-up time (median)b	Lag periodc	Outcomes used in meta-analysis	Ascertainment of (a) exposure (b) outcome	Reference population	Effect estimate
Pettersson, 1990	Sweden	Retrospective cohort	56 116	NR (conisation was the usual procedure; hysterectomy was carried out in relatively few cases)	CIN3	8.1 yearsb	1 year	Other cancers: corpus uterus, ovaries, breast	(a) and (b) Swedish National Cancer Registry	General Swedish female population	SIR
Bjorge, 1995	Norway	Retrospective cohort	37 001	NR (conisation was the usual Tx; alternatively, hysterectomy)	CIN3	9.1 yearsb	1 year	Other cancers: overall, corpus uterus, ovaries/fallopian tubes, colon/rectum, breast, lung/bronchus/trachea, female anogenital HPV-related (vagina, vulva, cervix)	(a) and (b) Cancer Registry of Norway	General Norwegian female population	SIR
Frisch, 1995	Denmark	Retrospective cohort	30 294	NR (some women might have received no treatment; hysterectomies might have been included)	CIN3	12.4 yearsb	No	Other cancers: lung	(a) and (b) Danish Cancer Registry	General Danish female population	SIR
Levi, 1996	Switzerland	Retrospective cohort	2190	NR	CIN3	10.1 yearsb	NR	Other cancers: overall, corpus uterus, breast	(a) and (b) Vaud Cancer Registry	General female population from Swiss canton of Vaud	SIR
Evans, 2003	UK	Retrospective and prospective cohort	59 519	NR (<10% had radical surgery; NR if all women had Tx but most probably did because only CIN3 cases were included)	CIN3	8 yearsb	No	Other cancers: overall, vulva, vagina, corpus uterus, ovaries, anus, colon/rectum, breast, lung, cervix/vagina, female anogenital HPV-related (vagina, vulva, cervix, anus)	(a) TCR (b) Prospective f-u (until 1982) for patients with CIN diagnosed before 1971; retrospective f-u through NHS Cancer Registry for patients with CIN diagnosed after 1970	General female population covered by TCR	SIR
Taylor, 2006	USA	Retrospective cohort	56 020	NR	CIN3	5 years	No	Other cancers: ovaries, lung	(a) and (b) California Cancer Registry	General female population from California	SIR
Edgren, 2007	Sweden	Retrospective cohort	125 292	NR (CIN has traditionally been treated by CKC, LC, cryosurgery, LLETZ; 5% were treated with hysterectomy)	CIN3	18.4 yearsb	1 year	Other cancers: vulva, anus, rectum, female anogenital HPV-related (vagina, vulva, anus)	(a) and (b) Swedish Cancer Registry	Women without previous history of CIN3	RR (adjusted for age, calendar period, socioeconomic status and parity)
McCredie, 2010	Australia and New Zealand	Retrospective cohort	72	CKC	CIN3	27.5 years	>∼2.5 yearsc	Other cancers: cervix/vagina Mortality: cervix/vagina	(a) Records of National Women’s Hospital (b) Medical records, histopathological review or cancer and death registries	General female population from New Zealand	SIR
Jakobsson, 2009 for mortality, 2011 for other cancers	Finland	Retrospective cohort	26 876 for other cancers; 25 827 for mortality	Excision (CKC, LLETZ, LC); ablation (LA, electrocoagulation, CT); other (other excision, cervix amputation etc.)	CIN1–3a	8.4 yearsb	f-u started at end of calendar year of CIN treatmentd	Other cancers: overall, vulva, vagina, corpus uterus, ovaries, anus, colon/rectum, breast, lung, cervix/vagina, female anogenital HPV-related (vagina, vulva, cervix, anus) Mortality: cervix	(a) National Hospital Discharge Register (b) Finnish Cancer Registry for other cancers; Finnish Cause-of-Death Register for mortality	General Finnish female population	SIR
Strander, 2007 for other cancers, 2014 for other cancers and mortality	Sweden	Retrospective cohort	132 493 in 2007; 150 883 in 2014	NR (the Swedish Cancer Register does not include date on treatment; hysterectomies have been included)	CIN3	17.5 yearsb in 2007; 20.9b in 2014	1 year	Other cancers: vagina (in 2007), cervix/vagina (in 2014) Mortality: cervix/vagina	(a) Swedish Cancer Registry (b) Swedish Cancer Registry for other cancers; Swedish Cause-of-Death Register for mortality	General Swedish female population	SIR
Saleem, 2011	USA	Retrospective cohort	124 075	NR (hysterectomies might have been included; NR if all women had Tx but most probably did because only CIS cases were included)	CIN3	NR	1 year	Other cancers: anus	(a) and (b) SEER registry (large population-based registry from 17 regions)	General female population covered by SEER registry	SIR
Gaudet, 2014	Canada	Retrospective cohort	54 320	NR (hysterectomies might have been included; NR if all women had Tx but most probably did because only CIN2+ cases were included)	CIN2–3	10.1 years	0.5 years	Other cancers: vulva, vagina, anus, female anogenital HPV-related (vulva, vagina, anus)	(a) British Columbia Cervical Cancer Screening Program (b) British Columbia Cancer Registry	General female population from British Columbia	SIR
Kirkegard, 2014	Denmark	Retrospective cohort	83 008	Cervical conisation (‘minor surgical procedure’, thus hysterectomies have probably been excluded)	NR (probably any CIN, histological or cytological)	14.9 years	No (lag period only for SIR of cancer incidence in the time window 1–5 years, but not for SIR of overall cancer incidence)	Other cancers: colon/rectum	(a) Danish National Patient Registry (b) Danish Cancer Registry	General Danish female population	SIR
Coffey, 2016	UK	Case-control study (nested case-control in the Million Women Study)	797 vulval cancers in a cohort of 1.3 million women aged 49–65 years; 19/797 had a history of CIN3	NR (hysterectomies might have been included; NR if all women with CIN3 had Tx but most probably did because the likelihood of expectant management of CIN3 is low)	CIN3a	13.8 yearsb,e	3 years	Other cancers: vulva	(a) and (b) UK National Health Service Central Registers (NHSCR) (self-reported data from the recruitment questionnaire were used to define most exposures, but NHSCR was used for ascertainment of CIN3)	Women with vulval cancer but no previous CIN3 diagnosis (case-control study)	RR (adjusted for smoking, alcohol, BMI, diabetes, age at menarche, oral contraceptive use, parity, prior tubal ligation, prior hysterectomy and deprivation)
Sand, 2016	Denmark	Retrospective cohort	156 290	NR (hysterectomies might have been included; some women might have received no Tx)	CIN2–3	13.6 yearsb	1 year	Other cancers: vulva, vagina, anus, female anogenital HPV-related (vulva, vagina, anus)	(a) Danish Cancer Registry & Pathology Data Bank (b) Danish Cancer Registry	Denmark population without history of CIN2/3	HR (adjusted for age and education)
Ebisch, 2017	The Netherlands	Retrospective cohort	89 018	NR (hysterectomies might have been included; NR if all women had Tx but most probably did because only CIN3 cases were included)	CIN3	14	1 year	Other cancers: vulva, vagina, anus, female anogenital HPV-related (vulva, vagina, anus)	(a) and (b) Dutch nationwide registry of histopathology and cytopathology (PALGA; Houten, The Netherlands)	Dutch population without history of CIN3	RR (adjusted for age)

BMI, body mass index; CIN, cervical intraepithelial neoplasia; CIS, carcinoma in situ; CKC, cold knife conisation; CT, cryotherapy; f-u, follow-up; HPV, human papillomavirus; HR, hazard ratio; LA, laser ablation; LC, laser conisation; LLETZ, large loop excision of the transformation zone; N, number; NR, not reported; RR, relative risk; SIR, standardised incidence ratio; TCR, Thames Cancer Registry; Tx, treatment.

^aSome women had cytological diagnosis (or not reported).

^bMean if median is not reported.

^cNo lag period, but we were able to exclude cancers occurring during the first 6 or 12 months after treatment (or no cancers occurred during the first 6 or 12 months).

^dThis means that the lag period varied from 0 to 12 months, depending on the month when treatment was carried out.

^eThis was a nested case-control from the Million Women Study and the mean reported is for a cohort of 1.3 million women aged 49–65 years.