Table 1. Description of included papers.

Study	Aim	Participant Sample Size at Baseline and Follow Up(s)	Study Location	Study Design	Follow up Period	Quality Rating	Effects Measure Reported in Study	Results	Calculated Unadjusted Odds Ratio (CI)	What Significance Test is Testing For	Measure of intention	Measure of contraceptive use
Curtis & Westoff 1996	To examine the relationship between stated ITU contraceptives and subsequent use during a three-year period	908 women married to same partner at both surveys, non- users at initial survey	Morocco	Longitudinal (cohort)	3 years	High (10)	Odds Ratio	OR: 6.78 *** aOR: 2.6 *** aOR (with interactions): 2.40	7.40 (5.51, 9.93)	Whether contraceptive use significantly increased among those reporting ITU compared to those not reporting	All ever-married respondents who weren’t using a contraceptive method were asked: “Do you intend to use a method to delay or avoid pregnancy at any time in the future/in the next 12 months?”	Not described
Lori et al. 2018	To examine the uptake and continuation of family planning following enrolment in group versus individual ANC	240 pregnant women at ANC settings at baseline and 164 at endline	Ghana	Longitudinal (cohort)	1 year	High (10)	Odds Ratio	aOR (any method): 1.549 aOR (any modern method):1.085	2.17 (1.11, 4.25)	Same as Curtis and Westhoff, 1996	Not described	Self-reported use
Sarnak et al. 2020	To assess the dynamic influence of unmet need on time to contraceptive uptake, as compared with that of contraceptive intentions and their concordance	747 sexually active, non-contracepting, fecund, women	Uganda	Longitudinal (cohort)	6,12,18, 24, and 36 months	High (9)	Hazard Ratio	HR: 1.65 * aHR: 1.45 *	3 years 4.48 (3.13, 6.42) 30 months 3.75 (2.62, 5.38) 24 months 3.22 (2.24, 4.62) 18 months 2.59 (1.79, 3.75) 12 months 2.27 (1.55, 3.33)	Same as Curtis and Westhoff, 1996	Non- contracepting women were asked whether they would use contraceptives in the future	Use of modern contraception
Tang et al. 2016	To (1) calculate the incidence of LARC use among postpartum Malawian women, and (2) assess if LARC knowledge and ITU LARC were associated with LARC uptake.	539 postpartum women (3 months), 480 (6 months), and 331 (12 months)	Malawi	Longitudinal (cohort)	3, 6, and 12 months after delivery	High (9)	Hazard Ratio	HR (implant use only): 1.88 ** aHR (implant use only): 1.95 *	1.05 (.67, 1.64)	Same as Curtis and Westhoff, 1996	Contraceptive methods she was planning to use in the first year after delivery	Self-reported use
Adelman et al. 2019	To evaluate which characteristics collected at the point of abortion are associated with contraceptive use over the extended postabortion period for women.	500 postabortion patients	Cambodia	Longitudinal (cohort)	4 and 12 months	Medium (7)	Odds Ratio	OR (4 months): 7.89 *** OR (12 months): 3.32 *** aOR (4 months): 4.60 *** aOR (12 months): 2.38	4.55 (3.00, 6.92)	Testing whether those who reported intention to use had different actual use compared to those who were undecided or reported they weren’t going to use a method	Not described	Self-reported use
Adler et al. 1990	To understand adolescent beliefs about contraception and their intention to use	325 postpartum, low-income, breastfeeding contraceptive initiators	USA	Longitudinal (cohort)	1 year	Medium (7)	Correlation coefficient	Pill (female): 0.42 *** Pill (male): 0.10 Diaphragm (female) 0.27 *** Diaphragm (male): 0.27 * Withdrawal (female): 0.20 ** Withdrawal (male): 0.46 ***	NA	Testing correlation of intention to use method with frequency of use in the following year	7-point scales responses to the statement "If I do have intercourse in the next year, I am ([very unlikely to very likely]) to ever use [method X] for birth control."	Self-reported use
Borges et al. 2018	To examine the effect of pregnancy planning status on the relationship between ITU and current use of contraceptives among postpartum women	474 ANC patients	Brazil	Longitudinal (cohort)	6 months after birth	Medium (6)	Concordance	28.9% concordance between contraceptive preference and subsequent contraceptive use.	1.48 (.54, 4.04)	Only assess significance by demographic or pregnancy planning group, not overall significance between ITU and contraceptive use	Women were asked while pregnant what type of contraceptive they intended to use after childbirth	Self-reported use and for those who reported more than one method, the most efficient was used.
Callahan & Becker 2014	To link women’s contraceptive uptake and experience of unwanted pregnancy between 2006 and 2009 to their unmet need status and their stated ITU contraceptives in 2006	3,933 married women at baseline and 3,687 at endline	Bangladesh	Longitudinal (cohort)	3 years	Medium (8)	Odds Ratio	OR (women with unmet need): 8.29 * OR (women with no unmet need): 7.17 *	7.25 (5.50, 9.56)	Same as Curtis and Westhoff, 1996	Pregnant and nonpregnant married women younger than 50 were asked: “Do you think you will use a method to delay or avoid pregnancy at any time in the future?” and were asked which method they intended to use	Self-reported use
Davidson & Jaccard 1979	To examine whether within versus across-subject procedures are more accurate for predicting behaviour from attitudes	279 married women at baseline and 244 at endline	USA	Longitudinal (cohort)	2 years	Medium (6)	Behavioural Intention B correlation	Correlation (for contraceptive use): 0.68 **	NA	Correlation between intention to use method and use within the next 2 years	7-point Likert scale measuring from likely to unlikely response to the statement: “I intend to use contraception within the next 2 years”	Self-reported use
Davidson & Morrison 1983	To understand factors that moderate the attitude-behaviour relation	221 married women, aged 18-38 years	USA	Longitudinal (cohort)	1 year	Medium (6)	Phi coefficients	Within and across subjects Condoms (within subjects): 0.86 ** Condoms (across subjects): 0.63 ** Pill (within subjects): 0.83 ** Pill (across subjects): 0.77 ** IUD: (within subjects): 0.94 ** IUD: (across subjects): 0.85 ** Diaphragm (within subjects): 0.92 ** Diaphragm (across subjects): 0.78 **	NA	Tests whether difference between within and across subject Phi-square coefficients is significant	Respondents intending to use a birth control method during the next year were asked what method they intended to use.	Self-reported use
Dhont et al. 2009	To investigate unmet need for LARCs and sterilization among HIV-positive pregnant women, and the impact of increased access to LARCs in the postpartum period on their contraceptive uptake	219 HIV-positive pregnant women at ANC settings at baseline and 205 at endline	Rwanda	Longitudinal (cohort)	9 months after birth	Medium (6)	Percentages	53% pregnant women reported an intention to use a LARC or to be sterilised after delivery 72% of women who had intended to start using a LARC actually did so at a site offering LARCs compared to only 4% of women at public FP sites ***	1.23 (.48, 3.21)	Tests whether LARC uptake at Site A (public FP services) were different than at Site B (guaranteed implant and IUD services)	Not described	Not described
Roy et al. 2003	To investigate women’s ITU a method as a measure of contraceptive demand	421 female participants in the 1992-92 National Family Health Survey	India	Longitudinal (cohort)	6 years	Medium (7)	Proportions	Of the 421 women who were asked the NFHS question on contraceptive intentions, 127 stated that they would use a method in the future. More than half (51%) of the women stating they would use a method in the future, did not do so during the intersurvey period compared to 29% of respondents who had said they would not practice family planning actually did so **	2.53 (1.53, 3.60)	Testing whether those who intended to use contraceptives were significantly more likely to use compared to those who had not planned on using a method	Not described	Self-reported use
Johnson et al. 2019	To understand how women’s prenatal infant feeding and contraception intentions were related to postpartum choices	223 postpartum women at baseline; 214 women postpartum in the hospital and 119 women at postpartum visit at <43 days	USA	Longitudinal (cohort)	Not specified	Low (5)	Correlation coefficient	Prenatal contraceptive intention and postpartum in-hospital correlation: 0.41 *** Prenatal contraceptive intention and postpartum visit choice correlation: 0.47 **	0.75 (.47, 1.22)	Correlation between prenatal contraceptive intention and in-hospital and postpartum visit method choice	Not described	For the analysis, contraceptive choice was characterized as no contraceptive method versus LARC

*p<.05, **p<.01, ***p<.001