The CAP Study
Women with a history of pre-eclampsia in their most recent pregnancy who agree to participate, receive calcium 500mg daily or placebo from enrolment until they become pregnant, and up to 20 weeks of pregnancy. All women receive calcium 1500mg daily after 20 weeks as per WHO recommendations. The primary outcome is pre-eclampsia.
CAP study progress
Of 603 completed pregnancies, 80 ended in miscarriage. We are on track to achieve our sample size of 540 women with completed pregnancies which progressed to 20 weeks or more, within the next 2 months. This should enable us to discuss the final results at the PRE-EMPT meeting in October.
Final data analysis plan
Thanks To Armando Seuc for tackling the complexity of this study in terms of intermediate outcomes:
One of the most important challenges of this study is the presence of various intermediate outcomes. Intermediate outcomes (IO) are in the causal path between treatment/intervention and final outcome; they differ from confounders and require specific statistical approaches .
There are three main types of IOs:
- They do not prevent us from observing the main outcome. For example, non-compliance with treatment does not prevent observing pre-eclampsia (neither pregnancy nor EPL). When the rate (of non-compliance) is important, we might be interested not in the “assignment effect” (ITT) but also in the “causal effect” (treatment receipt effect).
- They prevent us from observing the main outcome, but the main outcome does exist and can be imputed. For example, loss-to-follow-up (LTFU) (“missing”) does prevent us from observing pre-eclampsia (and/or pregnancy, early pregnancy loss, etc.)
- They prevent us from observing the main outcome, and the main outcome is not defined for some categories of the intermediate outcome.For example, pregnancy, because pre-eclampsia is not defined in non-pregnant women; for the same reason, early pregnancy loss (before 20 weeks’ gestation) is another example.
We will now describe how we will handle each of the above intermediate outcomes in this study:
- If non-compliance is small, it can be ignored. If not, we have two relevant effects, the “assignment effect” (ITT) and the “causal/receipt effect”; this can be estimated with Principal Stratification (PS) which keeps the advantage of the IV randomization.
- If loss-to-follow-up (LTFU) (“missing”) is small, it can be ignored. If not, sensitivity analysis is conducted, i.e. i) different rates of the main outcome pre-eclampsia (or pregnancy, etc.) are assumed for LTFU in each of the trial arms, ii) compound rates of pre-eclampsia are computed and iii) effect size and its precision of the difference (or ratio) between trial arms assessed.
- If the incidence of pregnancy (or EPL) is assumed to be independent of calcium supplementation, bias is not introduced when computing and comparing binomial rates (proportions) of pre-eclampsia among pregnant (and/or non-EPL) women between the trial arms. This independence can be assessed comparing i) the rates of pregnancy in the two arms, and ii) pregnant women in the two arms with respect to baseline characteristics. In this case comparison of pre-eclampsia rates in pregnant women between the two arms can be made using a straightforward chi-square test. We expect in this trial that pregnancy is independent of calcium assignment.
- If pregnancy (or EPL) is not independent from calcium assignment, we expect to observe differential rates of pregnancy (and/or EPL) in the two arms, but we cannot assume or impute differential rates of pre-eclampsia in non-pregnant (neither in EPL) because pre-eclampsia is not defined in these subgroups. It is then clear that sensitivity analysis does not make sense here. As already mentioned, it is very improbable that there is an effect of calcium on pregnancy, but this is not the case with respect to EPL and that’s why it makes sense (and it is convenient from a statistical point of view) to define “EPL and/or pre-eclampsia” as one of our primary outcomes; this seems to be the only sensible solution under the “simple” binomial rates approach, as in doing so we keep the benefits of randomization; however it might make results difficult to explain and/or interpret. Principal Stratification (PS) might be used as in (1) above, but there is no certainty it will work because we might have to make too many and unsustainable assumptions.
The above discussion makes us consider that survival analysis with competing risks should be applied to complement the results using the binomial rates approach as this technique is well equipped to handle the sequence of IOs already.
Archiving of biomarker study biological samples
The CAP study team and CoLab have collaborated with GSK Maternal and Neonatal Health Unit, Global Health R&D Unit, R&D Alternative Discovery & Development, who have measured a comprehensive panel of putative pre-eclampsia biomarkers on a sub-set of the CAP study population. The laboratory analyses in the UK are complete. Excess samples were to have been stored in the US, but we now have capacity for long-term archiving of the samples at the University of Stellenbosch biobank in South Africa, and arrangements are being put in place to do this.
The samples will be available for future pre-eclampsia research.
We have submitted for publication the manuscript for a methodological paper: Participant recruitment and retention in longitudinal preconception randomized trials: lessons learnt from the calcium and pre-eclampsia (CAP) trial.
1. Hofmeyr GJ, Novikova N, Singata M, Fawcus S, Oyebajo A, Munjanja S, Belizán JM. Protocol 11PRT/4028: Long term calcium supplementation in women at high risk of pre-eclampsia: a randomised, placebo-controlled trial (PACTR201105000267371) http://www.thelancet.com/protocol-reviews/11PRT-4028
2. Hofmeyr G, Belizán J, von Dadelszen P; the Calcium and Pre-eclampsia (CAP) Study Group. Low-dose calcium supplementation for preventing pre-eclampsia: a systematic review and commentary. BJOG. 2014 Mar 13. doi: 10.1111/1471-0528.12613
3. Seuc AH, Peregoudov A, Betran AP, Gulmezoglu AM. Intermediate outcomes in randomized clinical trials: an introduction. Trials. 2013;14:78.
4. Hofmeyr GJ, Lawrie TA, Atallah AN, Duley L, Torloni MR. Calcium supplementation during pregnancy for preventing hypertensive disorders and related problems. Cochrane Database Syst Rev. 2014 Jun 24;6:CD001059
5. Cormick G, Ciapponi A, Cafferata ML, Belizan J. Calcium supplementation for prevention of primary hypertension. Cochrane Database Syst Rev 2015 Jun 30;6:CD010037.
6. Hofmeyr GJ, Seuc A, Betrán AP, Purnat T, Ciganda A, Manyame S, Munjanja SP, Singata M, Fawcus S, Frank K Hall DR, Cormick G, Roberts J, Bergel E, Drebit S, von Dadelszen P, Belizan J on behalf of the calcium and Pre-eclampsia Study Group. The effect of calcium supplementation on blood pressure in non-pregnant women with previous pre-eclampsia: an exploratory, randomized placebo controlled study (WHO Study A65750). Pregnancy Hypertens. 2015 Oct;5(4):273-9
7. Hofmeyr GJ. Prevention of pre-eclampsia: calcium supplementation and other strategies. SA Journal of Obstetrics and Gynaecology, submitted
8. Cormick G, Betrán AP, Ciapponi A, Hall DR, Hofmeyr GJ; calcium and Pre-eclampsia Study Group. Inter-pregnancy interval and risk of recurrent pre-eclampsia: systematic review and meta-analysis. Reprod Health. 2016 Jul 18;13(1):83
- Hofmeyr GJ, for the Calcium and Pre-eclampsia (CAP) Study Group. Calcium supplementation for preventing pre-eclampsia/eclampsia. Africa Regional Meeting for Interventions for Impact in essential obstetric and newborn care. Addis Ababa 21-25 February 2011.
- Hofmeyr GJ, Singata M, Fawcus S, Belizan JM, Oyebajo AB, Munjanja SP, Sawchuck D, von Dadelzsen P, Bergel E, Betran AP, Purnat T, Tahuringana E.Long term calcium supplementation in women at high risk of pre-eclampsia: a randomized, placebo-controlled trial. 31st Conference on Priorities in Perinatal Care in Southern Africa. Kruger park, South Africa, 6-9 March 2012.
- Hofmeyr GJ, for the Calcium and Pre-eclampsia (CAP) Study Group. Use of calcium and vitamin D for prevention of pre-eclampsia/eclampsia. USAID MCHIP Asia Regional Meeting on Interventions for Impact in Essential Obstetric and Newborn Care, Dhaka, Bangladesh May 3-6, 2012
- Hofmeyr GJ, for the Calcium and Pre-eclampsia (CAP) Study Group. Low dose calcium supplementation for preventing pre-eclampsia: A systematic review and commentary. 32nd Conference on Priorities in Perinatal Care in Southern Africa, Mpekweni Beach Resort, 12-15 March 2013.
- Hofmeyr GJ. Calcium and Pre-eclampsia, Nutrition meeting, Birchwood, Htel Johannesburg, 14 April 2013
- Hofmeyr GJ. Evidence and global health: constraints and ethical dilemmas. African Cochrane indaba, Cape Town, 6-8 May 2013
- Manyame S, Hofmeyr GJ. Pre-pregnancy calcium supplementation for the prevention of pre-eclampsia. African Cochrane indaba, Cape Town, 6-8 May 2013
- Singata-Madliki M for the CAP study group. The effect of calcium supplementation on blood pressure in non-pregnant women with previous pre-eclampsia: an exploratory, randomized placebo controlled study (WHO Study A65750). 34th Priorities in Perinatal Care Conference, Champagne Sports Resort, Drakensberg, KZN 17-20 March 2015
- Hofmeyr GJ on behalf of the CAP Study Group.Calcium Supplementation in Pre-eclampsia for preventing maternal mortality. International Conference on Maternal and Newborn Health Research – a KLE Centenary Eveny, Belgaum, India 5-6 March 2016.
- Hofmeyr GJ on behalf of the CAP study group. The role of calcium in hypertension and pre-eclamsia. Cape Town cardiac disease in Pregnancy symposium, Hatter Institute for cardiovascular Research in Africa, Cape Town, 28 September 2016