Suppose that you’re at your doctor’s office, discussing an important health issue that may become a concern in the near future. There are multiple drugs available in the market that you can use to prevent unwanted outcomes. Some of them are so effective that there is practically no chance you will have a negative event if you start taking them. Effectiveness of the other options range from 94% to much lower, with the most commonly used drug failing about 10% of the time for the typical user. Somehow, you go home with the drug that has a one in 10 failure rate: worse, you’re not alone; most people end up in the same boat…
If that strikes you as implausible, it is not. This is where we are when it comes to contraceptive methods available to us: while LARCs (long-acting reversible contraceptives) – the “set and forget” methods of intrauterine device (IUD) and implant – protect against unwanted and mistimed pregnancies with close to 100% effectiveness and conveniently (without you having to do anything on a periodical basis), take-up rates for these methods, which have been around for a good while, are low everywhere in the world and close to negligible in the developing world, especially among adolescent females and young women. Of the minority of women who get counseling, most go home with condoms only or with condoms and the pill (which has an effectiveness rate of 91% as commonly used) or the injectable (DMPA, which has an effectiveness rate of 94% with typical use). You have to remember to take the pill everyday (preferably around the same time), while you need another injection every three months with DMPA (or DMPA-SC, the more recently introduced subcutaneous version that can be self-administered). Worse, many young women leave clinics without ever being informed and counseled about the more effective methods by their health provider: this is true not just in developing countries but in developed ones as well. What gives? Why are we putting up with unacceptably high rates of unwanted or mistimed pregnancies? Note that I am not talking about those who don’t use any modern contraceptive methods here (for whom we can debate issues of access and demand another time), but rather the choice of method for those on the intensive margin, i.e. women or couples who are using a method in order to regulate the number, timing, and spacing of their children.
Of course, the issue is more complex than I have described so far: effectiveness is just one aspect of these methods that matters to the women who are making the choice. All methods discussed above, hormonal or not, have side effects. The main one is a change in bleeding patterns, which can move in opposite directions: bleeding can increase and become heavier/prolonged or decrease, become lighter, lead to spotting, or disappear altogether (amenorrhea). Worse, these side effects are not only heterogeneous by product, but also idiosyncratically by individual. They are usually worse within the first few months after adoption and stabilize within 3-6 months. Another significant side effect for hormonal methods is the possibility of weight gain, along with perhaps less significant but still annoying effects such as headaches, acne, etc. These aspects need to be factored into the decision-making process, but given that no one method (other than condoms or less effective natural methods) has an obvious advantage over the others in these aspects, it seems unlikely that they would override the large advantage of LARCs with respect to effectiveness and convenience.
Discretion is another issue of importance for many women, who’d rather that other people, sometimes including their mothers or their partners/husbands, do not know about their choices regarding birth control. Here, LARCs do have an advantage (although rare cases of sexual partners complaining about feeling the strings of an IUD or the implant rod becoming visible in the arm of a thin person have been reported). Cost is another issue, although generally many methods are subsidized and the main difference is a higher up-front cost for LARCs, which end up being cheaper in the longer-run because of the long duration of effectiveness. So, credit constraints can cause individuals to opt for the pill, which may cost less than a dollar now (and every subsequent cycle), as opposed to a LARC that might cost $3-5 now, but does protect for at least 3-5 years. Finally, some people may find the idea of inserting a foreign (albeit quite small) object into their uterus or arm repellant: we all have our likes and dislikes and try to optimize along these dimensions.
Still, taking all of these factors into account, looking at the scientific evidence on each of these factors, it’s hard to imagine that the prevailing distribution of methods adopted is optimal – either from a private or social returns perspective. In other words, under a different scenario, we may have people choosing LARCs over non-LARCs more often, who can not only better plan the timing of their pregnancies but are better off, period. They may not only be better off in the short term, but also in the longer run when the knock-on effects of their increased control over fertility start becoming apparent with respect to human capital accumulation, labor market participation, and empowerment.
What would that world look like? Or, more importantly, how would we start heading in that direction? This is not an easy problem – just talk to people who have been working in this area for years, if not decades: many readily admit that we don’t know how to increase take-up rates of effective modern contraceptive methods, especially among adolescent females and young women. Issues on both the supply-side (provider training, provider bias, availability of products and stock-outs) and the demand-side (information, uncertainty, cost, social norms, etc.) affect take-up. Where does one start?
Recently, I have joined a team that began trying to tackle this issue in Cameroon. We first wanted to get a sense of the bottlenecks on both the supply- and demand-side, so that we could get some insights as to what interventions may be effective in unclogging some of the pressure points. I conclude this blog by highlighting a few of the messages that came out of our formative qualitative work conducted mainly with 15-19 year-old females, as well as our site visits, workshops, and brainstorming sessions with various experts in Cameroon.
- Emphasizing the long-acting nature of LARCs is counterproductive: Many adolescent females in Cameroon, as in other countries, want to delay pregnancy not for 5-10 years but for a shorter duration. So, when the first thing they might hear about a product is that it works for 3-5 (implants) or 10-12 (IUD) years, this fact is not only irrelevant, but may cause them to discount it quickly, especially when coupled with worries about these products causing infertility (being irreversible). Better to emphasize the convenience and effectiveness of these methods, emphasize that they can discontinue using them any time and quickly start trying to get pregnant, only to later mention that they can keep using this method up to X many years before renewing, switching, or discontinuing.
- When it comes to reversibility, people have the wrong idea: Perhaps given the long-acting nature of IUDs and implants, many young people are worried about how fast they would reverse. This concern is not as strong for the injectable or the pill. In fact, the truth is the opposite: the injectable has a slight delay in returning to fertility, compared with the other methods. This means that even if your time horizon for delaying pregnancy is short (say, less than a year), it may still make sense to adopt an implant (or IUD).
- Word of mouth is strong among teenagers and bad experiences with adopted methods can have negative spillover effects: Many of the interviewed mentioned side effects, particularly associated with the injectable (the Depo shot). What was interesting about the reports was that it was as much about personal experience with the product, as it was about recounting someone else’s experience as to why they themselves did not want to consider it. Ironically, such accounts were rare for LARCs – likely owing to their low take-up rates (most people don’t know about them or don’t have a friend or a relative who used them). This makes how these products are rolled out into new communities, sub-groups hugely important. If people are properly counseled about the side effects, know what to expect, what is normal and what is not; are actively followed-up to be reassured that their experiences are normal; and helped to manage discomfort or pain, the word of mouth among school-aged females or young mothers can be a positive. Hence, there are opportunities to increase demand, but only after the supply side is adequate.
- Adolescents do attend health clinics, but they almost never leave with a LARC: One might think that the problem of reaching adolescents is one of not being able to reach them at all. This is, of course, true to some extent: we could have more people receiving information and counseling about contraceptive methods in most settings. However, it is not true that adolescents are not at these clinics. But, they are rarely leaving with an IUD or an implant, especially if they’re nulliparous or unmarried. Thinking of ways to (a) increase the number of young people getting counseling and (b) making counseling more effective to diversify choice are potentially promising starting points to increase the uptake of LARCs.
- Speaking of counseling, there seem to be a number of deficiencies with the traditional methods of counseling: A nurse counselor may have 15-30 minutes at best for a typical counseling session, which is not enough time to discuss the relevant features of all available methods. Even if he/she did, behavioral science tells us that providing more information can actually be less effective than summarizing salient features of a smaller subset of suitable options in causing someone to make an active choice. It also tells us that there is no neutral framing. But, many counseling sessions do not discuss the most effective methods first while they try to cram too much into the short session. Visual aids, such as cue cards and posters, are similarly ill designed – emphasizing all features equally. One could imagine counselors identifying a couple of top options for the client that align with her stated preferences and circumstances and discussing those (first) with the help of carefully designed visual aids and decision-making tools.
- Speaking of behavioral biases, there may also be status quo bias (or path dependency) of method adoption. This study suggests that current users of injectables, wishing to delay pregnancy for more than three years, were less likely to adopt a LARC after being counseled about them. While this, by itself, is not solid evidence of path dependency, it makes complete sense that many people may be unwilling to deviate from their first adopted method even if a more effective and/or convenient method is now available and on offer. This suggests that it may be productive to encourage people to try the best methods first; actively follow-up their experiences and switch methods as desired/necessary. This may again increase welfare without any negative consequences, on average.
- On the supply side, problems are many: Knowledge of these products is inadequate even among providers. They clearly need better training on modern contraceptive methods, both theoretical, but especially practical, so that they are completely comfortable in inserting and removing LARCs, avoiding unpleasant experiences for the adopters. They also need to be well versed in all the side effects, medical eligibility issues, etc. They may need to spend more time following up with clients. Most importantly, they need to be effective counselors being able to help a client make an informed choice about what is an important and deeply personal decision steeped in judgments in her surrounding environment. Without knowledgeable, competent nurses, who are also effective and non-judgmental counselors, it is hard to see how any effort that tackles the demand side could be effective. In fact, increasing demand without adequate supply may well be counterproductive in the longer run.