The concept of unmet need

The development of the concept of unmet need is rooted in the KAP (Knowledge, Attitude, and Practice) surveys of the 1960s. Researchers identified married women whose preferences and behavior appeared contradictory—that is, they wanted to limit or space childbearing but were not using a method of family planning [11]. The KAP surveys gave way to the World Fertility Survey program in the 1970s and 1980s and ultimately evolved into the Demographic and Health Surveys (DHS) Program which covered a wider range of topics, starting in 1984. The algorithm for determining unmet need grew increasingly complex over time as questions on fertility preferences evolved and data from the contraceptive calendar were included when available. In 2012, a simplified, consensus DHS/MICS definition of unmet need was established using a standard algorithm [4]. A comprehensive history of unmet need and evolution of the classification schema has been detailed by other authors [12–14].

One concern with the concept of unmet need is that the term itself implies a demand for family planning—and, in fact, is summed with contraceptive prevalence to compute an indicator called “demand for family planning”—but the term does not necessarily reflect actual or potential interest in method use. In particular, it does not reflect how women themselves perceive their risk of pregnancy, the strength of their preferences, or their interest in or resistance to family planning. Additional concerns about the measurement of unmet need include the failure to differentiate married women who are sexually active from those who are not, and thus at no risk of pregnancy [12], the failure to include male partners [15, 16], and instability in professed fertility preferences [17–20].

The extent to which survey measures of unmet need gauge latent demand for family planning has been questioned due to the temporal instability of the measure and the number of different groups it encompasses [21]. Even in the early days of the development of unmet need as a measure of demand for family planning, it was known that in some countries less than half of women with unmet need were currently at risk of pregnancy [14]; unmet need classification depends on prospective fertility preferences as well as on ex-post facto assessments of the intendedness of pregnancies and recent births. A technical working group for FP2020 opted against using reductions in levels of unmet need as a global goal because it is not a unidirectional measure of programmatic success [22].

Despite concerns about its measurement and interpretation, unmet need is a powerful concept. Abortions, surreptitious use of family planning, and unwanted pregnancies all attest to an ongoing need for family planning that is unfulfilled [23]. How to assess women’s ‘need’ or demand for family planning has proven difficult, however. In particular, evidence indicates that—even among survey respondents whose fertility operates within the “calculus of conscious choice”—the answers to prospective fertility preference questions are fraught with ambiguity and uncertainty [17, 19]. Measures of unmet need depend on women’s tendency to plan and articulate fertility preferences in a two-year window. Fertility preferences are subject to both social context [5] and to vital conjunctures in women’s lives [24], including husband’s desire for children [16, 25], future economic well-being, marital stability, and survival of current children. Longitudinal evidence finds substantial instability in individual women’s fertility preferences over time [26–28].

Study context

At the time this study was being designed, Ghana had recorded one of the highest levels of unmet need for family planning among married women on the African continent, at 36 percent in 2008 [29]. Family planning use had declined slightly among married women, from 25 percent in 2003 to 24 percent in 2008. Meanwhile, Ghana’s Total Fertility Rate (TFR) in 2008 was among the lowest in West Africa, at 4.0 births per woman. Unmet need is typically only measured among currently married women, but the focus of this study is currently married and sexually active women combined, as both groups of women are at risk of unwanted pregnancies. Nationwide, 29 percent of married and sexually active unmarried women in Ghana had an unmet need for family planning as measured by data from the 2014 GDHS.

Ghana’s attainment of regionally low fertility despite modest levels of family planning use has been a demographic puzzle for nearly two decades [30]. Abortion is legal in Ghana and has been hypothesized as a reason for lower-than-expected fertility [31]; but evidence has been inconclusive. It may be that high levels of unmet need in Ghana partly reflected women’s growing tendency to articulate a need for spacing or limiting births. During early stages of the demographic transition, the percentage of women with unmet need can increase even as demand for family planning is being satisfied simply due to women’s increased interest in reducing fertility [32].

Ghana has a relatively strong family planning program. The contraceptive method mix is diverse. Injectables, the pill, and implants are the most common methods, followed by the rhythm method. Women can obtain contraception from public and private sources. Family planning is inexpensive but not free. Ghana does experience occasional contraceptive supply issues and there are some limits to the method mix offered. Social marketing campaigns have proven successful but some very remote areas of the country remain a few hours’ distance from the nearest clinic. Even so, in surveys women rarely cite access and cost as reasons for non-use of family planning [33].

By 2014 the TFR increased slightly, to 4.2 births per woman [34]. The 2014 GDHS also found an increase in modern contraceptive prevalence since 2008 (from 17 to 22 percent) and a decline in unmet need (from 36 to 30 percent) among married women ages 15–49. This brings the country on par with levels of unmet need in neighboring West African countries, but still high in a global perspective.

Perceptions about side effects and attitudinal factors pose a challenge to increased family planning use in Ghana. Focus group discussions from a hospital in Ghana found women’s concern with menstrual regularity results in dissatisfaction with family planning methods that prevent menstruation [35]. Another study found that many Ghanaian women perceive family planning as ineffective or unsafe [36], and DHS data from 1988 to 2008 show that attitudinal resistance has been an increasing component of unmet need in Ghana [33]. Male attitudes toward contraception are mixed: in the 2014 GDHS, 73 percent of men age 15–59 rejected the idea that contraception is a woman’s business and men should not have to be involved, but 46 percent supported the statement that women who use contraception may become promiscuous [34]. Married Ghanaian women’s sexual empowerment is a statistically significant predictor of contraceptive use, even after controlling for other factors [37].

Ethical considerations

The ICF International Institutional Review Board (IRB), which requires compliance with the U.S. Department of Health and Human Services regulations for the protection of human subjects (45 CFR 46), reviewed and approved all study procedures and questionnaires. A waiver of written consent was obtained from the IRB due to minimal risk of harm and a lack of procedures for which written consent is normally required. Respondents were asked for verbal consent to be re-contacted during the main survey and for verbal consent to be interviewed and to be audiotaped at the start of the follow-up interview. Before the interview began interviewers were required to provide their electronic signature attesting that they had received verbal consent from the respondent to be interviewed and that they had correctly indicated whether the respondent consented to be audiotaped.

In keeping with IRB regulations and The DHS Program’s practices, the confidentiality of the respondent’s information was maintained at all stages of the survey. Recordkeeping used anonymous cluster and respondent identifiers. As voluntary HIV serotesting was conducted during the 2014 Ghana DHS, no data entry on names or addresses was done and all cluster and household numbers were scrambled prior to linkage with HIV test results. Similarly, at the conclusion of fieldwork the implementing agency for the follow-up study destroyed all identifying information used to contact respondents, maintaining only an anonymized identification number. A linkage between the anonymized identification number for the follow-up study and the final, publicly available dataset is kept only by The DHS Program.

Sample characteristics

The 96 married and sexually active unmarried women analyzed in this article were systematically selected from among GDHS respondents. The follow-up sample was intended to be diverse, but given the scale of the study, the sample was not designed to be perfectly representative of the three regions. Table 1 indicates how the characteristics of the sample as defined in the GDHS compared with family planning users and with women who had an unmet need in the three selected regions. GDHS sample weights are applied to regional percentages and to the sub-study respondents. Table 1 shows that respondents to the follow-up study were more concentrated in their 30s than their respective regional counterparts. Fewer follow-up respondents were age 15–19. Follow-up respondents were more predominantly rural than both family planning users and women with unmet need in the country as a whole.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Background demographic characteristics of follow-up respondents compared with married and sexually active unmarried women in three selected regions, GDHS 2014.

https://doi.org/10.1371/journal.pone.0182076.t001

Women who reported using family planning in the GDHS were more highly educated and wealthier than women with unmet need. Both follow-up samples were over-representative of the lowest wealth quintile than the regional averages. The follow-up sample exhibited religious diversity, but was more heavily traditional/spiritualist and no stated religion than women nationwide in both the unmet need and family planning groups. One of the clusters in the North was Konkomba-speaking, and respondents were ethnically Gurma; the follow-up study sample was thus much more heavily comprised of Gurma women than the country as a whole.

Reproductive characteristics of the four groups (regional and follow-up respondents who were family planning users and who had unmet need in the GDHS) are shown in Table 2. Women’s knowledge of family planning methods is high in Ghana. Nationwide, over 99 percent of women with unmet need know at least one modern method. In the follow-up survey, one respondent was identified by the GDHS as not knowing any method of family planning, another as only knowing traditional methods. The majority of women with unmet need have used a family planning method before.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Background reproductive characteristics of follow-up respondents compared with married and sexually active unmarried women in three selected regions, GDHS 2014.

https://doi.org/10.1371/journal.pone.0182076.t002

Fertility preferences

Fertility preferences are a pivotal component of unmet need. Among fecund women, declared intention to have a/another birth and the preferred timing of the next birth determine unmet need status. The two questions on reproductive preferences used to compute unmet need for this group are: (1) “Would you like to have (a/another) child, or would you prefer not to have any (more) children?” Allowable responses to this question are: (a) want a/another; (b) no more; (c) cannot get pregnant; (d) a special condition, such as ‘after marriage’; or (e) don’t know/undecided. Non-pregnant respondents who want a/another child are then asked: (2) “How long would you like to wait from now before the birth of (a/another) child?” Pregnant women are asked: “After the child you are expecting now, would you like to have another child, or would you prefer not to have any more children?” The answer must either be a specific number of years and months, a special wait condition, or undecided. While the questions used to ascertain fertility preferences are seemingly straightforward, women’s ability and willingness to articulate a fixed timeline for their preferred time to next birth are culturally and temporally variable.

Non-pregnant respondents who want no more children and meet criteria for fecundity and non-use are considered to have an unmet need for limiting. The consensus definition of unmet need uses a threshold of two or more years to determine whether women have an unmet need for spacing. A comparison between responses to the question from the GDHS and the follow-up survey are shown in Table 3. As the table indicates, 18 percent of follow-up respondents with unmet need and 13 percent of follow-up family planning users gave inconsistent answers to this question between surveys. The shift in responses occurred largely among women who stated they were undecided in the GDHS but said they wanted another child upon follow-up. There was some additional negligible movement in both directions between wanting no more children and wanting another child during the follow-up.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Comparison of preference for a/another child as reported to the GDHS and in follow-up survey.

https://doi.org/10.1371/journal.pone.0182076.t003

The main ambivalence in fertility preferences that emerged through interviews was not whether to have another child, but the timing of that preference. In the GDHS, responses to question about desired timing of the next birth were limited to a single number of months or years, or to a special wait condition. In the follow-up survey women were allowed to specify an open range of desired time until the next birth. Responses to this question, grouped by minimum wait time into one-year intervals, are shown in Table 4 for follow-up respondents with unmet need. Figures include the eight respondents who said in the GDHS that they were undecided or that they wanted no more children but declared in the follow-up survey that they wanted a/another. Twenty of 31 respondents indicated a desire to wait until a single point in time, while the remainder gave a time range, averaging 23 months.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Preferred minimum waiting time to next birth among follow-up respondents with unmet need, according to follow-up survey.

https://doi.org/10.1371/journal.pone.0182076.t004

Of the respondents classified as having unmet need who indicated in the follow-up interview that they wanted a/another birth, 11 declared in follow-up a minimum desired waiting time of less than two years. For seven of these 11 respondents, two years encompassed both the minimum and maximum waiting time, which would have classified them as having no unmet need. The other four respondents gave a range of time that started before two years. If they had been asked to give a fixed number and settled on less than two years, they would also have been excluded from the group defined as having unmet need.

The follow-up survey also asked women the strength of their desire to wait that long for the next birth. Results are shown in Table 4. Desires to delay were weakest among women who expressed a preference to wait one to three years; they tended to be strongest at the highest and lowest time boundaries.

A main theme that emerged from the discussion of fertility preferences overall was that respondents felt torn between joy versus means—or in many cases torn between the potential future value of a child (who could be the “star of the family”) versus the immediate cost of raising a child. Respondents and their partners valued children highly, but most respondents perceived an inherent trade-off between the value of another child and the financial or health costs of too many or closely spaced births. For example, as a 35-year old woman in urban Central Region (R03.08) described:

Well, it would be good to have another child, and it would be especially good to have a safe delivery. Even the Bible tells us that children are a blessing and a gift from God, so if I have another child and am able to take care of him or her such that he grows to become responsible and respectable person, it would even bring honor to me. People would point at him and say, oh there goes Sister [Name]’s son, and that would bring me fulfillment and joy… I do not know how far the child might go in life; he might even be an important personality and bring honor to our family, so that could be a value that having lots of children might bring us… Even if you look at the Bible, it says to be fruitful and multiply and to replenish the earth, it’s out of many children that some grow to be important personalities of the world but it’s due to economic and financial hardships that people face that’s why they may decide to have fewer children but ideally, you should have more, you never know which of your children would be of significance someday.

Respondents frequently indicated that they would accept an unwanted pregnancy but would not be happy. “I don’t like it but if it comes I will accept it” or “I will not be happy but once it has come I will accept it” was how seven respondents (R01.01, R01.03, R01.05, R02.03, R02.07, R04.02, R05.07) described their attitude toward having another child. “I will be disturbed but I will be happy to have a child” said another (R01.07).

For the 11 respondents identified as having an unmet need whose timing shifted below the two-year threshold, including those who had originally said no more, two themes emerged. First was an overall ambivalence toward pregnancy, not simply conflicted feelings but also a sense of fatalism. The majority of women in this category were in rural areas and had parity of four or more. The second theme that emerged was that the respondent had a partner who would be happy or very happy about another birth while the respondent herself did not feel strongly. For example, one 40-year old rural respondent in the Northern Region with seven children (R13.02) who changed from wanting no more children to wanting a child now indicated the decision was her own, but that she was deferential to her husband’s preference.

I know [my husband] likes children even though we have nothing. Take a look at my house, we don’t have strength but we are managing with life… We talk all the time. He said we are going through a tough time but do we still want more children or we should try a method? And I said whatever you decide I have no problem with it… I would like to have another child any moment from now.

In sum, 41 of 50 respondents with unmet need consistently answered the question about desirability of a/another child in the GDHS and the follow-up. The main source of inconsistency was movement from undecided to wanting another child. Including respondents who changed from wanting no more children to wanting a/another child, there were 11 cases where the stated time to next birth moved into the two-year range, seven of which (14% of unmet need sample) would have meant that women were not classified as having unmet need.

Discrepancies in reporting of method use

At the beginning of the follow-up survey, after questions to confirm key identifying characteristics, non-pregnant women were asked if they were currently using family planning. The wording of the question was the same as in the DHS (“Are you currently doing something or using a method to delay or avoid getting pregnant?”). Unlike in the DHS, however, women who responded “no” were prompted about natural methods (“What about the rhythm/calendar method? What about withdrawal?”).

A comparison of method use reported in the GDHS and the follow-up survey is shown in Table 5. Out of 44 non-pregnant respondents that the GDHS determined to have unmet need for family planning, 15 (34 percent) reported in the follow-up interview that they actually were using a method of family planning. Five reported using a modern method, nine reported a traditional method, and one reported both. Additionally, three respondents who said they were using a method in the GDHS reported not using a method in the follow-up study.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 5. Comparison of current family planning use, as reported to the GDHS and follow-up survey.

https://doi.org/10.1371/journal.pone.0182076.t005

Respondents who reported discrepant use of a method in the GDHS and the follow-up survey were closely examined for identity verification questions and checked against other women in the household to see whether the follow-up interview may have been conducted with the wrong household member. None of the GDHS respondents with unmet need who reported using a method at the time of follow-up lived in a household with another woman who used a method. All matched on at least seven of nine pieces of identifying information.

Some of the women may have started or stopped a method in the intervening days between interviews. In cases where women were asked about the discrepancy, none reported starting between interviews. Two users of modern methods said that they were using a method and had told the GDHS interviewers but it was not recorded. The DHS question on method use is specifically intended to be inclusive of traditional methods (“doing something or using a method to prevent pregnancy”) and in the DHS interview, immediately prior to the question on current method use, respondents are asked whether they have heard of a list of methods, including rhythm or calendar, withdrawal, and lactational amenorrhea. They are read a brief description of each method type. But the main theme that emerged among respondents who reported natural methods was that they misinterpreted the GDHS question about method use as being about modern methods. For example, here is an exchange with a 26-year old woman in an urban part of Central Region (R05.06):

Interviewer: I’d like to begin by confirming the information I received. Are you currently doing something or using any method to delay or avoid getting pregnant?

Respondent: No, I am not on any medication.

Interviewer: Yes, I understand you may not be on any medication to prevent pregnancy but there are other ways of preventing pregnancy such as the rhythm or withdrawal methods. Are you currently using any of these to prevent pregnancy?

Respondent: Oh, yes, I didn’t understand at first. I use the withdrawal method.

Interviewer: Please according to the information I have from the previous interviewer, you are currently not using any method or means to prevent pregnancy. Is that information correct?

Respondent: Yes, it is. When they came they only asked about the modern methods of family planning. They didn’t ask about the rhythm and the withdrawal like you did. That is why I said no but my husband and I use the withdrawal.

Four other respondents reported similarly that they understood the question as being about modern methods. As a 33-year old woman in a rural part of Central Region (R04.05) stated: “When they asked this question, they asked about the family planning, they didn’t explain that the methods included rhythm and withdrawal, that’s why.” Overall, of the ten women who had been classified by the GDHS as having unmet need but who reported natural methods, eight initially answered “no” to a repeated GDHS question about method use and only revealed natural method use after they were asked about rhythm and withdrawal.

Abstinence

Unmet need for family planning is designed to be computed among women who are exposed to the risk of unintended pregnancy. Exposure to the risk of pregnancy is assessed in two stages: first through self-reported sexual activity, and second through estimated fecundability. As mentioned, the definition of unmet need can apply to sexually active unmarried women as well as married women. By definition, all women who are currently in union are assumed to be at risk of pregnancy [4]. The obvious concern about this assumption is that a non-negligible proportion of women who are currently in union report long spells of sexual abstinence, for example, due to spousal migration or postpartum abstinence. A recent analysis finds that, along with fear of side effects, infrequent sexual activity is one of the two most common reasons cited for not using family planning among women with unmet need in Africa, Asia, and Latin America [7]. If all women who cited infrequent sex or abstinence as a reason for non-use were excluded from the definition of unmet need, the overall level of unmet need would decrease by a country average of 16 percent [12].

Apart from the underreporting of the rhythm and withdrawal methods, one additional type of omission that emerged from follow-up interviews was the intentional use of prolonged abstinence, particularly during the postpartum period, in order to space births. While the practice of postpartum abstinence has long been known as a dimension of birth spacing in Central and West Africa [44], and abstinence is the most effective way to prevent pregnancy, prolonged abstinence is not typically considered as a traditional method of family planning by surveys such as DHS. Respondents are asked about the date of last sexual activity separately from questions about methods to prevent pregnancy.

What is missing from DHS assessments of abstinence but emerged from the follow-up interviews is the intentionality behind abstinence. In their follow-up interview, 8 of 44 non-pregnant respondents indicated that abstinence was intentionally ongoing and was designed to prevent pregnancy. For example, one respondent, a 36-year old woman in a rural part of Northern Region (R12.05) said “if he sleeps with me I will be pregnant, and I don’t want to be, and he is okay with the decision [to be abstinent].” Others cited lengthy durations of postpartum abstinence for the purposes of birth spacing. “The gods we worship do not allow [family planning] so when you have a small child you avoid sex” was how a 20-year old respondent (R10.03) described it. A 31-year old rural woman in Northern Region (R10.05) responded, “We discussed that the child is small so we should wait for the child to grow first and because of that, I don’t sleep in the same room with my husband again… my husband said he will not touch me until my child grows.”

Abstinence within marriage is typically considered a temporary state that does not preclude unmet need, as the woman may be at risk of unwanted pregnancy as soon as the abstinence ceases. Indeed, some women indicated that they were abstinent because their husbands were traveling, or that they were withholding sex because of womanizing; such abstinence is involuntary and, based on their fecundability, these respondents will be ‘in need’ of a family planning method as soon as sex resumes. But the most recurrent reason women cited for abstinence was that it was intentionally designed to avoid pregnancy, typically to allow time for their youngest child to grow, and thus unlikely to cease soon: in other words, abstinence was being used as a method of family planning.

Abortion

The GDHS did not ask about abortion, but follow-up respondents were asked if they had ever had an abortion, and if so how many times. Underreporting of abortion is always a concern in surveys. The follow-up study prefaced the abortion question with phrasing used in prior surveys: (“Women sometimes take steps to end their pregnancy, for example if they find themselves pregnant when they do not want to be, or when it is difficult for them to continue with their pregnancy because of opposition from their husband, partner, relatives or others.”) Then, instead of asking whether women had deliberately ended a pregnancy, we asked using local terminology: "Have you ever been in a situation where you or someone else have had to 'put your hand' to your pregnancy?" If needed, interviewers clarified that they were asking about abortion. The majority of respondents reported never having had an abortion: 10 percent of women with unmet need and 28 percent of GDHS family planning users reported having had an abortion, mostly only once. The pooled result, that 19 percent of all respondents reported having ever had an abortion, is broadly consistent with the 2007 Ghana Maternal Health Survey, which found that 15 percent of all women reported having had an abortion in their lifetime [45].

Reasons for non-use of family planning

In discussing use and non-use of family planning, five themes emerged: gaps in knowledge, fear of side effects, cost and access, the role of male partners, and religion.

Cost and access.

Follow-up respondents were asked about perceived cost and access barriers to obtaining contraception. Table 6 displays the results on cost and access barriers among non-users compared with confirmed users. Here we can see that 15 percent think modern contraceptives are difficult to access and 16 percent think they are expensive, compared to 12 and 10 percent respectively among users. The more important difference is in the don’t know category: an additional 16 percent of non-users don’t know if contraception is difficult to access, most because they have never tried to obtain it, and 38 percent don’t know if it is expensive.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 6. Comparison of perceived cost and access barriers to contraception among contraceptive users and non-users.

https://doi.org/10.1371/journal.pone.0182076.t006

As indicated in Table 6, 15 percent of respondents with unmet need perceived access barriers to contraceptives. Access barriers include not only distance to health facility, waiting time, and availability of supplies, but also operating hours: a study in the Nkwanta district of Ghana found that favorable opening hours was the most significant access factor [47]. Most respondents mentioned that contraception was available in hospitals, clinics, and pharmacies. “Every pharmacy sells some. Immediately you mention what you want, they will give it for you unless they don’t have some,” said one 35-year old respondent in Greater Accra (R06.02). Another 40-year old respondent in rural Central Region (R02.07) said, “They can easily be accessed from the hospitals and the pharmacy shops at a low cost.” The only concerns about access tended to be in rural Northern villages. “There is no way you can access the service in this village. We have to travel to the clinic which is far from here, to get the service” (Age 37, rural Northern Region, R13.12).

Family planning is not free in Ghana, but it tends to be offered at very low cost. Among confirmed non-users of modern methods, 38 percent were unaware of cost or had had no opinion on it. Only one respondent explicitly cited a cost barrier. In general, most respondents with unmet need felt that expense did not pose a barrier. As a 41-year old woman in rural Central Region (R02.01) said, “For something that you can use to avoid pregnancy and space your child births, it should have been really expensive, but it has been made in such a way that everyone can afford to use it.” At the same time, this woman said she preferred traditional methods.

Consistent with the GDHS, in the follow-up study the majority of respondents with unmet need reported knowing a source of family planning. In general, contact with health care facilities had a positive or somewhat neutral influence on non-users. Women reported having been educated through their contact with fieldworkers and health providers. Cost and access emerged in follow-up interviews, but not as a central theme for non-use.

Overall comparison of reasons for non-use.

The GDHS survey asked most women with unmet need about their reasons for non-use in an open-ended way. (“You have said that you do not want (a/another) child soon [or: You have said that you do not want any (more) children.] Can you tell me why you are not using a method to prevent pregnancy? Any other reason?”) Nationwide in 2014, among married and sexually active unmarried women classified as having unmet need who were asked why they were not using a method of family planning, 49 percent stated that fear of side effects or health concerns was a reason for non-use. (Note: Multiple reasons are allowed; women who are pregnant or who have undefined fertility preferences are not asked this question.)

Toward the end of the follow-up survey, women who confirmed that they were not using family planning were asked the same open-ended question. For the 30 women who were asked the GDHS question about non-use, reinterviewed for the follow up survey, and confirmed as non-modern method users, I examined the follow-up transcript for reasons for non-use, including the repeated question when applicable. Table 7 shows the results of this comparison by reason type. Each reason and its count from the GDHS are shown. On the right-hand side, the table shows the percentage of cases where that theme was independently confirmed as a reason for non-use of family planning. I also show the number of respondents who indicated that theme as a reason for non-use in the follow-up survey but had not mentioned it in the GDHS survey. On average, respondents gave 1.1 reasons for their non-use of family planning in the GDHS but 3.0 reasons in the follow-up; the latter estimate would have been higher if respondents who reported actually using a method were excluded.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 7. Correspondence between reasons for not using family planning in the GDHS and follow-up.

https://doi.org/10.1371/journal.pone.0182076.t007

The follow-up study was interested in independently understanding unmet need; no attempt was made to quiz respondents on the validity of the reason they gave to the GDHS for non-use. The results shown in Table 7 indicate that fear of side effects, a major theme from the GDHS, was mentioned as a reason for non-use among 23 of 30 respondents. Opposition from husband or partner and religious opposition also appeared to have been underreported in the GDHS. Meanwhile, despite additional prompting on cost and access, no additional cases where cost or access posed a barrier were found.

Some respondents gave reasons not mentioned in the GDHS that are included in Table 7. Two were explicitly fatalistic about non-use of family planning (“we are under the care of the gods so am not using anything to protect myself” said a 20-year old respondent in the rural Northern part of Ghana, R09.06). Two said they needed more information; one planned to start family planning soon. Seven of the 30 women interviewed had revised their fertility preferences within the two-year window, and five others (R05.02, R05.04, R06.01, R09.02, R09.06) included a statement along the lines of “I just don’t want it,” or “My heart does not like it.” Reliance on abortion did not emerge as a theme for any respondents.

Study limitations and strengths