How do you measure access to family planning?

By Rebecca Oas, Ph.D. | March 2, 2016

“Achieve, by 2015, universal access to reproductive health”

-Millennium Development Goals, Target 5B

Since the mid-1990s, the dominant rationale for family planning groups has been to provide strictly voluntary access to products and services, a shift away from the fertility-reduction focus that gave rise to infamously coercive programs. But how exactly do you measure access? The most frequently used measurement from the International Conference on Population and Development era (1994 onwards) and the Millennium Development Goal era (2000-2015) was “unmet need” for family planning. But “unmet need” does not contain any measurement of access to contraceptives or intent to use them, despite the false assertions of many family planning advocates. Furthermore, scholars studying family planning continue to call for a unified definition of access that can be used in international fertility surveys, but have not yet arrived at any firm consensus.

At a glance, it doesn’t seem as if it should be too difficult. Existing surveys already have measurements of what percentage of women with “unmet need” for family planning attribute it to lack of access: that’s 4-8% of women with a purported “need,” which comes to no more than 2% of married women overall in the regions with the greatest “need.”

scale of unmet need

(Data from Sedgh and Hussain, Studies in Family Planning, 2014)

Only 1-6% of women with “unmet need” claim a lack of knowledge of contraceptive methods, which is a dramatic decline from 25% twenty years ago. So women around the world are aware of family planning, and most of those not using it, despite wishing to avoid or postpone pregnancy, are making a conscious choice in the matter. So why not define access based on women’s own perception of it?

To put it more generously, family planning advocates prefer a broader concept of “access”—one that sees women’s religious beliefs or beliefs about the health risks of contraceptive use (“myths” that often turn out to be true) as barriers to access rather than legitimate factors in decision-making. To be less generous, family planning advocates are ultimately far more interested in increasing use of contraceptives than increasing access, but recognize that access is a much better term for public relations purposes.


New goals, new indicators

“By 2030, ensure universal access to sexual and reproductive health-care services, including for family planning, information and education, and the integration of reproductive health into national strategies and programmes”

            Sustainable Development Goals, Target 3.7

With the launch of the new Sustainable Development Goals (SDGs), statisticians at the UN are busily preparing a new batch of indicators to measure progress. Family planning groups are lobbying for a new indicator to replace “unmet need” as a proxy measurement for access. According to the latest report from the UN’s expert group on SDG indicators, the new proposed family planning indicator has been brought on board, albeit with one important change: they use the word “need met” instead of “demand satisfied,” which I would argue is a good thing.

The proposed indicator in its current form:

Proportion of women of reproductive age (aged 15-49 years) who have their need for family planning satisfied with modern methods

So why is this an improvement on “unmet need”?

When a form of this indicator was initially proposed for the SDGs, the arguments in its favor were:

This indicator is an improvement over the MDG Indicator on unmet need because it is more easily understood and is linearly correlated with contraceptive prevalence.

Let’s take those one at a time. The idea that it’s easier to understand than “unmet need” is a bit disingenuous because if you look at the calculation used to measure it…

                                   [contraceptive prevalence]

Need satisfied =          _______________________________

                                        [contraceptive prevalence] + [“unmet need”]

…you see that “unmet need” is an essential component of its definition. You do not get a simpler concept by taking a confusing concept and adding additional complexity. So, there goes that argument.

What about that linear correlation with contraceptive prevalence? Given that contraceptive prevalence exists in both the numerator and the denominator of the new indicator, the correlation is definitely pretty linear:

need satisfied graph 1

(In this and the following graphs, each dot represents a country. Percentages are of married women surveyed.  Data are from the UN’s Department of Economic and Social Affairs, 2015)

The question is why this is necessarily a good thing. For the sake of comparison, let’s plot “unmet need” against contraceptive prevalence:

need satisfied graph 2

Not so linear, especially at lower levels of contraceptive prevalence. What does this mean? Simply put, that you can have low levels of “need” in countries with low contraceptive prevalence—a fact that the new indicator neatly obscures with its aesthetically-pleasing linear correlation. As I’ve argued before, this new indicator makes it difficult for countries with low contraceptive prevalence to perform well. But why is that the case?

Consider this: the population of married women responding to national fertility surveys fall into three categories, where A+B+C=100%:

A) Women using modern family planning methods (contraceptive prevalence)

B) Women described as having an unmet need for family planning (unmet need)

C) Women not using modern methods and with no desire to avoid pregnancy (no need)

The proposed indicator takes A and B into consideration, but not C. What happens if we plot the percentage of women with no “need” for family planning against the new “need satisfied” indicator:

need satisfied graph 3

That’s a fairly linear inverse correlation: the more women with no “need” for family planning, the worse a country will appear to be doing in terms of satisfying said “need.”

Ellen Starbird of USAID has been a big proponent of the new proposed indicator, saying that it’s “a measure of what women are doing, rather than what women should be doing.” But Starbird has also called for “strong outcome indicators” with a proposed benchmark of 75% of “need satisfied.” If “need satisfied” is linearly correlated with contraceptive prevalence, then how is that benchmark not a quota for contraceptive use by another name?


Whither access?

The fundamental flaw in the old “unmet need” family planning indicator is that doesn’t treat women choosing not to use contraceptives—even if they would prefer, all other things being equal, not to get pregnant—as having made a valid choice. The new proposed “need satisfied” indicator retains that flaw and adds a new one: by tying the measurement of “satisfaction” to contraceptive prevalence, it disregards the presumably satisfied women with no need, desire, or demand for contraceptives and effectively penalizes countries for having too many of them.

Ultimately, this indicator is supposed to be somehow related to access to family planning services, yet the measurements being proposed are taking us further and further away from any meaningful reflection of that. So here’s an idea: we already have a measurement of lack of access as a subset of “unmet need.” The Guttmacher Institute analyzed the underlying causes of “unmet need,” including lack of access, in 51 developing countries. What would happen if we were to use self-reported lack of access instead of “unmet need” in the new “need satisfied” indicator?


                                  [contraceptive prevalence]

Need satisfied =          _______________________________

                                         [contraceptive prevalence] + [“unmet need”]


MY MODIFICATION (in red below):

                                  [contraceptive prevalence]

Need satisfied =             __________________________________________­­­­­­­­­­­­­­­­­­­

                                            [contraceptive prevalence] + [self-reported lack of access]

need satisfied graph 4

Not one of the 51 countries would have a measurement of “need satisfied” below 90% if this were really about access. Contrary to Starbird’s assertion, this proposed indicator is not about what women are doing, or what they want to be doing: it’s all about what family planning groups think they should be doing. As a strategy to market contraceptives and get the world’s governments to foot the bill, this indicator succeeds brilliantly. For pretty much anything else, it’s profoundly unsatisfying.