Gates Open Res Gates Open Research 2572-4754 F1000 Research Limited London, UK 10.12688/gatesopenres.13442.1 Research Article Articles The relationship between shared-decision making in contraceptive counseling and satisfaction in Accra and Kumasi, Ghana

[version 1; peer review: 2 approved with reservations]

 Compton Sarah Conceptualization Formal Analysis Funding Acquisition Investigation Methodology Project Administration Supervision Writing – Original Draft Preparation Writing – Review & Editing https://orcid.org/0000-0002-1539-5397 a 1 Manu Adom Investigation Methodology Project Administration Writing – Review & Editing 2 Maya Ernest Investigation Methodology Project Administration Writing – Review & Editing 2 Morhe Emmanuel Investigation Methodology Project Administration Writing – Review & Editing 3 Dalton Vanessa Conceptualization Writing – Review & Editing 1 Obstetrics and Gynecology, University of Michigan, Ann Arbor, MI, 48109, USA University of Ghana, Accra, Ghana Kwame Nkrumah University of Science and Technology, Kumasi, Ghana sarahrom@umich.edu

No competing interests were disclosed.

 20 12 2021 2021 5 180 8 12 2021 Copyright: © 2021 Compton S et al. 2021 This is an open access article distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Background: Current use of modern methods of contraception remain low in many parts of sub-Saharan Africa, including Ghana. One way to improve both satisfaction with and continuation of contraceptive usage is to increase the level of shared decision-making around method choice. In this study, we sought to evaluate the extent to which patients in urban Ghana experienced shared decision-making and if this was associated with method chosen, satisfaction, or continuation of the method at three-months post-visit.

Methods: We conducted a longitudinal survey. Women were recruited when they were starting a new method of contraception and followed-up with at three-months post-initiation from five family planning clinics in Accra and Kumasi, Ghana. Participants were asked who made the decision about their method choice, the patient herself, the provider, or the patient and provider together. Our outcomes included measures of satisfaction and three months’ continuation.

Results: Fifty-eight percent of our participants reported making the decision of which method to use themselves, and eighty percent reported being satisfied to be leaving with their chosen method. At three months, those who reported they engaged in shared decision-making were more likely to report they would choose the same method again (p=.003), a measure of satisfaction. Patients who reported they made the decision of which method to use (p=.002) and those who left with an injection or pill (p=.019) rated their provider less favorably, while participants who had used a method before (p=.024) and those who reported they received their method of choice (p=.000) rated their providers more favorably.

Conclusions: Measured in multiple ways, women who made the decision of which method to use were less satisfied. These results show the importance of providers engaging with patients during the contraceptive decision-making process.

 Contraceptive counseling Ghana Shared decision-making Gates Foundation OPP1170991 This work was supported by the Gates Foundation OPP1170991. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Introduction

Despite concerted efforts by the government of Ghana and its development partners to increase contraceptive access and use, current use of modern contraception remained below targets in both the 2008 and 2014 Ghana Demographic and Health Surveys ( GSS, 2009; GSS, 2015). In the 2017 Maternal Health Survey, 25% of women of reproductive age were using a modern method of contraception ( GSS, 2018). Unmet need for contraception in the country is currently estimated to be 35% for married women and 42% for sexually active unmarried women ( Staveteig, 2016). This is in spite of nearly universal knowledge of contraception, and very few cost and access barriers to use. Further, ever-use of contraception is much higher, suggesting that women may have low levels of satisfaction with available methods.

One way that has been explored in the United States to improve patient satisfaction with the medical services they receive is to adopt shared decision making techniques ( Shay & Lafata, 2015). Shared decision making is a form of health communication that emphasizes that there are multiple courses of care that would be appropriate to solve the problem. The provider offers his or her medical knowledge while the patient contributes his or her preferences and a decision is reached together ( Charles et al., 1997). In this way, the patient and the provider work together to choose a course of action which is medically sound and which meets the patient’s preferences.

Contraceptive decision-making is one which is perfectly suited to utilize a shared decision making framework, as the majority of women have multiple methods which are medically appropriate available to them ( Elwyn et al., 2014). Also, since methods have very different modes of action, efficacy, characteristics, and side effects, methods are not universally preferred by women ( Lessard et al., 2012). While shared decision making has begun to be described and studied in the US ( Dehlendorf et al., 2014a), and has been identified by the World Health Organization as an integral part of family planning counseling ( WHO, 2005), it has not been investigated in low- and middle-income countries (LMIC). In LMIC settings, providers are encouraged to use tools such as flip charts to help engage patients in contraceptive decision-making ( WHO, 2005), although it is not clear the extent to which providers use this or other tools, or to what extent these tools encourage shared decision-making.

In this study, we aimed to explore the extent to which patients in Ghana’s two largest cities (Accra and Kumasi) experienced shared decision-making and if this was associated with method chosen, satisfaction, or continuation of the method at three-months post-visit.

 Methods Sample and procedures

We used data from the Identifying Method-Related Factors Associated with Contraceptive Discontinuation in Ghana study, which has been described in other published work ( Rominski et al., 2018). The inclusion criteria for this study included women over the age of 18 who were starting a new method of contraception. They were recruited at the time of their counseling session, and interviewed both immediately before and after their contraceptive counselling in a private room adjacent to the family planning clinic. Women who met the inclusion criteria were identified to trained research assistants by clinic staff and were approached and invited to participate in the study. Those who said they were interested were taken through a comprehensive verbal consent process. As many women in this setting do not read well, it was determined by the ethics board that a comprehensive verbal consent process would ensure proper consent was obtained. All data were collected via interview-administered survey, using a tablet computer with Qualtrics (Provo, UT) offline software, but could be collected using Open Data Kit (ODK), an open-source software for collecting, managing and using data in resource-constrained environments.

Pre- and post-visit, as well as three-month survey records were matched using phone numbers, either the participant’s, or that of another family member, including her husband. As phone calls were the means of follow-up over the study period, no participants were enrolled who did not have a phone number.

Data from participants who left their initial visit with a method and were able to followed up with at three-months were included in the analysis.

Measures

In the pre-visit survey, participants were asked about their age, level of education, marital status, previous pregnancies, previous use of contraception, and preferences for contraception, including, “at this time, which method do you most prefer?”.

During the post-visit session, participants were asked, “during this visit, who made the decision about which method of contraception you would use?”, with the response options of, the provider, mostly the provider, the provider and me together, mostly me, me. These were collapsed into three responses for analysis; provider-driven (“the provider” and “mostly the provider”), shared decision (“the provider and me together”), and patient-driven (“mostly me”, and “me”).

In order to understand patient satisfaction with both the method of contraception with which they were leaving, and the process of counseling, patients were asked during the post-counseling both, “how satisfied are you to be leaving with this method?” as well as to rate the provider they saw on a series of 11 statements including, “respecting me as a person”, “considering my personal situation when advising me about birth control”, and “telling me the risks and benefits of the birth control method I chose”. The full set of questions can be seen in the appendix. These were initially collected from 1, for poor, to 5 for excellent. These were each dichotomized with those answering very good or excellent coded as 1, and those answering poor, fair, and good coded as 0. These were then summed to create a single patient assessment of their provider score, which would theoretically range from 0 to 11, and used as the outcome variable in a linear regression.

To further examine levels of satisfaction with their method, during the three-month phone call, participants were asked if they would choose this method again, and if they would recommend this method to a friend. Also in the three-month phone call, participants were asked if they were still using the method they had adopted at their visit, and those who answered no were asked the reason or reasons why they had stopped using it.

 Analysis

We used cross-tabs with Chi Square analysis to identify associated factors. Linear regression was used to determine factors associated with patient assessment of their provider. All analyses were conducted in Stata (Version 13.1, StataCorp, College Station, Texas) but could be conducted in R, a free-to-use analysis software.

All study procedures and documents were reviewed and approved by the Ethical Review Committee of the Ghana Health Service and the Institutional Review Board of the University of Michigan (HUM00129703).

Patient and Public Involvement: Patients and the public were not involved in the design, conduct, reporting, or dissemination of this research. Healthcare providers at study sites were involved in the design and conduct of the surveys and the results have been reported back to them.

 Results

A total of 405 participants, of the initial 537 who were enrolled, met the inclusion criteria for this analysis; leaving their counseling session with a method, and being followed-up with at three months (response rate: 75.4%). The average age of our participants was 28.5 years, with a standard deviation of 6.4 years. The majority reached junior secondary school or less (n=254, 62.7%), were married (n=224, 55.3%), and almost all (96.8%) have been pregnant at least once. Choice of method was dominated by long-acting reversible (LARC) methods (implants and IUDs) and the injection. Most participants reported making the decision of which method to use themselves, and most reported being satisfied to be leaving with their method. For more information on the sample, see Table 1.

Participant characteristics.
Variable N (percentage)
Age Mean (SD) 28.5 (6.4)
Highest level of education None 52 (12.8)
Primary school 47 (11.6)
Junior secondary school 155 (38.3)
Senior secondary school 97 (24.0)
More than secondary school 53 (13.1)
Missing 1 (.2)
Married Yes 224 (55.3)
No 171 (42.2)
Missing 10 (2.5)
Number of previous pregnancies 0 13 (3.2)
1 81 (20.0)
2 83 (20.5)
3 66 (16.3)
4 56 (13.8)
5 45 (11.1)
6 23 (5.7)
7 or more 38 (9.7)
Number of live births 0 19 (4.7)
1 99 (24.4)
2 107 (26.4)
3 68 (16.8)
4 53 (13.1)
5 21 (5.2)
6 or more 14 (3.5)
Number of abortions 1 88 (21.7)
2 37 (9.1)
3 or more 25 (6.2)
Ever used a method before Yes 242 (59.8)
Contraceptive method chosen at this visit
Implant 173 (42.2)
IUD 37 (9.1)
Injection 147 (36.3)
Pill 12 (3.0)
Sterilization 3 (.7)
None 21 (5.2)
During this visit, who made the decisions about what birth control method you would use?
Provider-driven 16 (4.0)
Shared decision-making 57 (14.1)
Patient-driven 236 (58.3)
Missing 96 (23.7)
How satisfied are you that you are leaving with this method?
Dissatisfied 14 (3.6)
Neutral 58 (14.3)
Satisfied 321 (79.3)
Missing 12 (3.0)
My level of participation in the decision was…
Not enough 1 (.2)
Just right 212 (52.3)
Too much 181 (44.7)
Missing 11 (2.7)
Did you receive the method you wanted? Yes 352 (86.9)
No 42 (10.4)
Missing 11 (2.7)

Of the total 405 participants from the five study clinics who were able to be followed-up with at three months, 350 (86.4%) were still using their method, and this differed by method. While all of the participants who chose female sterilization were still using this method at three months, as would be expected, 71.4% (n=105) of the participants who chose one of the injections, and 83.3% (n=10) of those who chose the pill, were still using their method at three months. Continuation rates were higher for the LARC methods with 94.3% (n=35) and 96.3 % (n=166) still using the IUD and implants respectively at three months. ( Figure 1)

Continuation at three-months by method.

While 87.2% of participants who said they received their method of choice were still using it at three months, 76.2% of those who said they did not receive their method of choice were (p=.051). Further, 92.2% of those who said they would choose the method again were still using it, while 72.7% who would not choose their method again were still using it (p<.001). Similar differences in proportions were found in whether the participant would recommend their method to a friend ( Figure 2).

Satisfaction and decision-making among participants still using their method at three-month continuation.

While there was no difference in continuation at three-months by who made the decision of which method to use, and the majority of participants reported at three-months post visit that they would choose their method again, this differed by who made the decision of which method to use. While 85.7% of the participants who engaged in shared-decision making would choose this method again, 75% of those who reported the provider made the decision, and 62.4% of those who made the decision themselves would choose the method again, and this difference is significant (p=.003) ( Figure 3).

Proportion of women still using their method and who would choose their method again at three months by who made the decision of which method to use.

Finally, using the participant rating of their provider immediately post-counseling as the outcome variable in a linear regression, participants who reported they made the decision of which method to use rated their provider significantly lower than others. On average, all else equal, patients who reported they got the method they wanted scored their provider 2.88 points higher on the 11-point scale (p<.001), and those made the decision of which method to use themselves scored their provider 1.63 points lower (p=.002). Further, those who left with either the injection or the pill rated their provider 1.06 points lower on the scale (p=.019), and participants who have used a method before scored the provider .990 points higher (p=.024). There was no significant difference in those who were still using at three months, or by age (although these approached significance in bivariate analysis and so were retained in the multivariate model). ( Table 2)

Linear regression with patient assessment of provider as the outcome variable.
Variable Beta Significance (p-value)
Got preferred method 2.88 .000
Patient decision -1.63 .002
Injection or pill -1.06 .019
Used method prior .990 .024
Still using 3 months -.616 .352
Age -.003 .932
Discussion

Discussion. In this study of urban Ghanaian women who were adopting a new method of contraception, 86.5% reported they were satisfied to be leaving their counseling session with their method, and 83.8% said they were leaving with the method they wanted. The majority of participants reported making the decision themselves about which method to use, however, as measured in multiple ways, those who did so were less satisfied, and rated their provider lower than those who reported either making the decision with the provider, or when the provider made the decision. These results show the importance of providers engaging with patients during the contraceptive decision-making process. The fact that women who made the decision themselves were the least satisfied, as measured in multiple ways, is important given the current model of contraceptive counseling where the provider gives information or education about each method, but leaves the decision up the patient ( Dehlendorf et al., 2014b; Upadhyay, 2001). Counseling in contraception has been criticized in the past for being too directive and not respecting patient autonomy, and therefore, providers have perhaps been encouraged to be non-directive in their counseling. The findings presented here that patients who make a decision with their provider are more satisfied, and rate their provider more favorably, suggests that providers can engage more actively with their patients to elucidate their preferences and help match those preferences with a method.

While we did not find an association with continuation at three-months as has been found in other investigations ( Dehlendorf et al., 2016), patient experience of the care they receive is an essential component of health care quality ( Berwick et al., 2008), and three months may be too short of a follow-up time to see differences in adherence.

This study has a few limitations. The relatively small sample size, coupled with the fact that these are only urban women, limits our ability to generalize to the larger Ghanaian population. Relying on patient assessment of their provider, rather than observing and objectively rating the providers is also a limitation, although patient experience with their provider is inherently subjective, and is an important facet of their own experience. The fact that almost one-quarter of our participants did not answer “during this visit, who made the decisions about what birth control method you would use?” may bias our outcome.

 Conclusion

Engaging in shared decision-making is associated with improved satisfaction among clients in urban Ghana. The women in this study were more satisfied with their contraceptive counseling session when their providers engaged in shared decision-making rather than making the decision for the patient, or allowing them to make the decision on their own. This study supports the development and implementation of shared decision making interventions for contraceptive counseling in this setting. Rather than shying away from engaging with patients and helping them decide which method of contraception to use, providers in this setting can be assured patients may be wanting this sort of counseling.

 Data availability Underlying data

Harvard Dataverse: Replication Data for: Shared-decision making in contraceptive counseling and satisfaction in Accra and Kumasi, Ghana, https://doi.org/10.7910/DVN/Y8UCIB. ( Compton, 2021)

Data are available under the terms of the Creative Commons Zero "No rights reserved" data waiver (CC0 1.0 Public domain dedication).

 Berwick DM Nolan TW Whittington J : The triple aim: care, health, and cost. Health Aff (Millwood). 2008;27(3):759–69. 18474969 10.1377/hlthaff.27.3.759 Charles C Gafni A Whelan T : Shared decision-making in the medical encounter: what does it mean? (or it takes at least two to tango). Soc Sci Med. 1997;44(5):681–92. 9032835 10.1016/s0277-9536(96)00221-3 Compton S : Replication Data for: Shared-decision making in contraceptive counseling and satisfaction in Accra and Kumasi, Ghana.Harvard Dataverse,2021. http://www.doi.org/10.7910/DVN/Y8UCIB Dehlendorf C Krajewski C Borrero S : Contraceptive counseling: best practices to ensure quality communication and enable effective contraceptive use. Clin Obstet Gynecol. 2014a;57(4):659–73. 25264697 10.1097/GRF.0000000000000059 4216627 Dehlendorf C Kimport K Levy K : A qualitative analysis of approaches to contraceptive counseling. Perspect Sex Reprod Health. 2014b;46(4):233–40. 25040686 10.1363/46e2114 4487742 Dehlendorf C Henderson JT Vittinghoff E : Association of the quality of interpersonal care during family planning counseling with contraceptive use. Am J Obstet Gynecol. 2016;215(1):78.e1–9. 26827879 10.1016/j.ajog.2016.01.173 Elwyn G Dehlendorf C Epstein RM : Shared decision making and motivational interviewing: achieving patient-centered care across the spectrum of health care problems. Ann Fam Med. 2014;12(3):270–275. 24821899 10.1370/afm.1615 4018376 Ghana Statistical Service (GSS), Ghana Health Service (GHS): Demographic and Health Survey 2008.Ghana: Accra, Ghana: GSS, GHS, and ICF Macro,2009. Reference Source Ghana Statistical Service (GSS), Ghana Health Service (GHS), ICF International: Ghana Demographic and Health Survey 2014.GSS, GHS, and ICF International, Rockville, Md, USA,2015. Reference Source Ghana Statistical Service (GSS), Ghana Health Service (GHS), ICF: Ghana Maternal Health Survey 2017.Accra, Ghana: GSS, GHS, and ICF,2018. Reference Source Lessard LN Karasek D Ma S : Contraceptive features preferred by women at high risk of unintended pregnancy. Perspect Sex Reprod Health. 2012;44(3):194–200. 22958664 10.1363/4419412 Rominski SD Manu A Maya E : Contraception method preferences among women in two Ghanaian cities. Glob J Reprod Med. 2018;6(1):555677. 10.19080/GJORM.2018.06.555677 Shay LA Lafata JE : Where is the evidence? A systematic review of shared decision making and patient outcomes. Med Decis Making. 2015;35(1):114–31. 25351843 10.1177/0272989X14551638 4270851 Staveteig S : Understanding Unmet Need in Ghana: Results from a Follow-up Study to the 2014 Ghana Demographic and Health Survey.DHS Qualitative Research Studies No. 20. Rockville, Maryland, USA: ICF International,2016. Reference Source Upadhyay UD : Informed choice in family planning: helping people decide. Popul Rep J. 2001; (50):1–39. 11552404 World Health Organization, Department of Reproductive Health and Research: Decision-making tool for family planning clients and providers; a resource for high-quality counselling.2005; Accessed 16th March 2019. Reference Source 10.21956/gatesopenres.14699.r36048 Reviewer response for version 1 Nelson Anita L 1 Referee Department of Obstetrics and Gynecology, College of Osteopathic Medicine, Western University, Pomona, USA

Competing interests: Grants/Research: Daré Bioscience, Viatris Pharmaceuticals Inc, Organon & Co, Co Ltd, Sebela PharmaceuticalsHonoraria/Speakers Bureau: Mayne Pharma, Organon & CoConsultant/Advisory Board: Bayer, Exeltis USA, Inc, Mayne Pharma, Sumitomo Pharma America

 12 4 2024 Copyright: © 2024 Nelson AL 2024 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve-with-reservations

Major Comments

I deeply appreciate the magnitude of the problems faced by researchers in settings such as this to obtain meaningful results that may help guide policy designed to help reduce unintended pregnancy rates and improve maternal-child health. I applaud the authors for attempting this work, but I do have some issues with how the study was conducted, how the results were analyzed, and what conclusions were drawn. Some of my concerns may be answered by reanalyzing the data already collected; others may just have to be acknowledged as limitations in the study design.

Fundamentally I am perplexed by the finding that it was possible that a woman who reported that she received the method that she herself had decided she wanted to use was not satisfied to be leaving the clinic with that method. Do the authors have any insights into that blatant incongruity? This dichotomy seems to profoundly impact the outcomes of the study and needs to be addressed.

I am also unclear about how to interpret the woman’s responses to the question about who decided the method she got. How do the authors think the woman who had decided upon a method before she entered the clinic and the provider agreed to her choice would categorize that? Would the woman claim credit for her choice or would she categorize it as “shared decision making”?

I am also not clear about how the healthcare system worked. The authors report the contraceptive method chosen at this visit. However, they do not report if the patient received the method, if she actually had started the method and how much supplies did she receive. Did women who chose IUDs have them placed at the visit or did they have to return later for that service? How many packs of pills were women given? Would we expect them to run out of supplies in 3 months and have challenges in returning for more supplies to be able to continue the use of that method? If would help international readers if the authors could describe how their system works beyond making the claim that “there are very few costs and access barriers to use”.

Abstract

It is troubling that the authors seem to claim as true that which they propose to study. They state that shared decision-making would increase method continuation and satisfaction. They should say that it may increase those desired outcomes. The results section switches outcomes at 3 months from wanting to use the same method to rating their providers at visit discharge. This is confusing. Find a theme rather than doing a data dump.

Introduction

Is the description of shared decision-making in this section what was asked in the survey? If not, then perhaps the authors could describe this vision as an idealized version and describe what it is that they mean in this study.

Methods

The authors do not report how many potential subjects were excluded because of not having telephones. This would be important to know to be able to generalize the findings of this study. Since the phone number given could be that of the male partner, were there any women who had confidentiality concerns that kept them from participating?

It is not clear exactly by stating that subjects were those who were “starting a new method”. Could it be that the subject had used the method in the past? Was the success of that prior use controlled for in the analysis? Subjects were apparently only asked if they had ever used a method before.

Were women asked about pregnancy plans? Could that have affected method selection? Were those who planned/desired pregnancy in the near term excluded?

I think it would be useful to display the full distribution of the answers to the question of who made the decision about what method. I am not happy with the collapse. The “mostly me” did have some provider input which is being ignored in this analysis. The authors refer to that group as “made the decision themselves,” which is not correct.

I am also troubled that the category of “good” was not added to “very good” or “excellent” but was combined with “poor” and “fair”. That seems potentially biased. Excellence should not be the enemy of good. At a minimum, the authors should do a sensitivity analysis to investigate if reclassifying those “good” responses would change the results.

The authors report that they asked those who were not using their method at 3 months why they had discontinued the use of the method, but I did not see an analysis of those answers. They may have had nothing to do with satisfaction or counseling. Readers need to know so they can interpret the continuation outcomes.

Results

Were there any differences between those who were enrolled and provided information at the initial date but were not included in the 3-month follow-up analysis and those who were able to be followed? Were there demographic differences or method differences? To lose 25% of the study population could profoundly affect the outcomes.

Why were women who underwent permanent contraceptive procedures included in a continuation study? Their numbers are too small to provide any meaningful information. They should be deleted, and this paper should concentrate on users of reversible contraceptive methods. Similarly, 12 subjects who used pills out of a total of 405 seems to be too small to subgroup and should certainly not be analyzed for continuation rates. The only way they could have gotten a higher rate is if all users continued use at 3 months.

It would be good to see what proportion of those who were still using each method at 3 months would choose it again. Figure 3 should be displayed by method. This would answer the concerns skeptical readers might have that the reason for higher continuation rates among IUD and implant users was not due to an inability to gain access to providers to remove the devices. In other words, that the reported higher continuation rates were not being coerced by systemic barriers.

Perhaps it is the wording, but it would help to understand how it would be that women would decide to use methods that they did not want to use. Is there something in the wording that created this disconnect? Are there supply problems? Are women started on one method in anticipation that they would quickly switch to another when supplies arrived or when the woman was at a different time in her cycle?

I actually was relatively impressed that women had such great autonomy in selecting the method that they were not being coerced into the use of methods or at least being directed by clinicians into “choosing” a method not to their liking. The problem there seems to be that clinicians do not offer any counseling or information to patients. Is it possible that others on the team have that responsibility? Possibly a community worker, a health education, or a nurse?

Discussion and conclusions

The claim that women were less satisfied if they made their own choice will have to be re-visited if the categories are re-grouped and the very small groups are deleted (permanent contraception, pill users).

Table 2 – When it says, “used method prior”, it represents the question on Table 1 “Ever used a method before”, and does not imply prior experience with the method she initiated at the enrollment visit, right? That should be clarified on the label or in a footnote on that table. Limitations should include that only urban women with anticipated ongoing access to a telephone were eligible. The missing information in many of the questions could easily have added bias but based on the results collected, these findings may be considered hypothesis-generating and warrant future investigation.

I would be happy to review any revisions the authors may choose to submit in the future.

Is the work clearly and accurately presented and does it cite the current literature?

Partly

If applicable, is the statistical analysis and its interpretation appropriate?

Partly

Are all the source data underlying the results available to ensure full reproducibility?

Yes

Is the study design appropriate and is the work technically sound?

Partly

Are the conclusions drawn adequately supported by the results?

Partly

Are sufficient details of methods and analysis provided to allow replication by others?

Yes

Reviewer Expertise:

Contraception

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

 10.21956/gatesopenres.14699.r35909 Reviewer response for version 1 Stevens Rose 1 Referee https://orcid.org/0000-0002-6858-4765 University of Oxford, England, UK

Competing interests: No competing interests were disclosed.

 29 2 2024 Copyright: © 2024 Stevens R 2024 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve-with-reservations

Summary of article

This study in urban Ghana primarily investigates the relationship between self-reported shared decision making during a contraceptive counselling consultation and continued use of contraception, intention to use a method again, and intention to recommend the method to a friend using chi-squared tests. Secondarily, it looks at the impact of several variables on a composite score of questions about the satisfaction with the provider’s actions/information during the consultation using linear regression. The authors find, amongst other things, that women who reported shared decision making during their contraceptive consultation showed the highest proportion of women reporting that they would choose their method again and those who reported making their contraceptive decision themselves showed the lowest proportion. They conclude that women may benefit from having consultations from providers that are more involved in contraceptive decision making.

Major comments

Overall, I think that investigating different measures of method satisfaction and seeing how provider involvement in decisions can impact this, as well as the impact of other dimensions of provider quality of care, is an interesting and worthwhile question. However, I think that it is difficult to robustly trust the conclusions of this analysis given the high proportion of missing responses in the main share-decision making variable, given potential selection biases from including only those who chose to use a method and were contactable on follow up, and given that much of the analysis is unadjusted for potential biasing confounders. With some additional analysis and reporting, and careful discussion of potential for bias, I think the conclusions should be reconsidered and checked for whether they are still considered robust. Additionally, I would encourage the authors to adjust the framing of the article away from ‘non-use of contraception’ as a problem and instead consider the problems of ‘dissatisfaction with contraception’ or ‘lack of contraceptive autonomy to use a preferred method’ as a more rights and justice-based framing.

Minor comments

Abstract

 

Methods – please include sample size in methods.

Results - personally, I would prefer to see fewer results reported to leave space for including effect sizes or differences in proportions, and confidence intervals to understand the magnitude of effect of different measures of satisfaction.

Results – “At three months, those who reported they engaged in shared decision making were more likely to report they would choose the same method again (p=.003), a measure of satisfaction” – can you specify as compared to what? I assume those who decided themselves, but it could be those who felt that the provider decided for them. Similarly, what is the comparison of those who left with the pill or injection? Those that left with nothing, or maybe implants? In general, would be best to focus on reporting the most salient results with more detail I think, rather than so many.

Introduction

 

“Despite concerted efforts by the government of Ghana and its development partners to increase contraceptive access and use, current use of modern contraception remained below targets in both the 2008 and 2014 Ghana Demographic and Health Surveys (GSS, 2009; GSS, 2015).” – Aggregate contraceptive use targets have been widely critiqued as reducing women’s contraceptive autonomy and increasing risk of contraceptive coercion. Setting up missed current use targets as the ‘problem’ that this article is interested in solving diverts the attention away from patient satisfaction as the main outcome. Personally, I would advocate for considering lack of satisfaction and counselling support, or the presence of contraceptive worries as valid issues in and of themselves as the problems instead. This is particularly interesting in the case of urban women in Ghana who have been shown to be worried about the impacts of contraception, show low satisfaction with modern contraception, and instead turn to a myriad of fertility regulation strategies to achieve their own reproductive goals. Some useful refs here would be: (Ref 1);  (Ref 2); (Ref 3); (Ref 4); (Ref 5)

2. “While shared decision making has begun to be described and studied in the US (Dehlendorf et al., 2014a), and has been identified by the World Health Organization as an integral part of family planning counselling (WHO, 2005), it has not been investigated in low- and middle-income countries (LMIC).” – This surprises me. It is not my area of expertise but I would have guessed there is some literature on shared decision making in contraceptive counselling in LMICs. Would love to see a little more review/discussion of studies on shared decision making in FP counselling in LMICs, or if not, a couple of sentences on what decision making looks like in the consultations in LMICs and Ghana typically.

Methods

1. Check consistency of counselling/counselling spelling.

2. “Data from participants who left their initial visit with a method and were able to followed up with at three-months were included in the analysis.” How many individuals left the consultation without a method and how many were unable to be followed up at 3 months? Could you comment on how this might impact the estimates of satisfaction and the characteristics of your sample (e.g. those most dissatisfied with their counselling may be the least likely to uptake a method.) I note many of your sample have been pregnant at least once. I wonder if the inclusion criteria for those who took up a method in their consultation disproportionately excludes women without children?

3. I’m not wholly convinced of the validity of summing the different domains of provider approval into one score out of 11 when they represent quite conceptually different questions. I would like to see a table or list of all 11 questions included as part of the main manuscript rather than the appendix, so that readers can assess the questions’ ability to be defined as one concept themselves. I can see that broadly they would all be indicative of better quality of care from the provider, so it is in some ways an interesting outcome. To be honest, I think the questions cited here are very interesting and would make for an interesting analysis when analysed individually e.g. looking at the impact of whether ‘“considering my personal situation when advising me about birth control’ impacted whether participants would choose their method again (perhaps a future paper focused on just these, and remove these results from the current paper).

Additionally, on this measure, I’m not entirely sure how robust it is to model it linearly given that it is effectively a count. Did you do any checks of model fit/diagnostics to check this approximation worked for your data? How well does the data fit a normal distribution and would a different distribution fit better? Also, please give details on how linear regression was used. Was each factor modelled individually in separate models or combined in one model? There is reference to bivariate and multivariate models later on but no reference to this in the methods.

Results

1. Table 1: Contraceptive method chosen at this visit: I believe this does not add up to 405, are some missing? Also there is a “None” option. In the methods it says “Data from participants who left their initial visit with a method and were able to followed up with at three-months were included in the analysis.” Were women who did not select a method included in this 405 or not?

2. Table 1: Some of the other variables also appear not to total 405 which intuitively should, e.g. number of live births. Please include a missing row for these or correct as appropriate.

3. Table 1: I noticed that 23.7% did not answer the shared-decision making question. Given that this is the key exposure variable of the study, please could you comment on why so many were missing and how this may bias the analysis.

4. Table 1: I find it very interesting that ~45% reported that their level of participation in the decision making was too much. Was this explored qualitatively at all or can the authors comment in the discussion thoughts as to why this may be given the local context?

5. In terms of the relationship between shared decision making and choosing to use the method again, I would be wary of strong conclusions without any adjustment for potential confounding variables. If women who are more likely to label themselves as having made the decision alone are also those with demographic characteristics associated with being wary of contraception and the health system for instance, they may be least likely to choose the method again. Ideally, this analysis would also include a multivariable regression analysis adjusting for conceptually relevant confounders, such as age or whether they have children. Directed acyclic graphs could be used to consider potential biasing pathways.  

6. Table 2: Please give confidence intervals for the beta estimates and include in the legend how variables were selected for or retained in the model. Is it one model for which each factor is mutually adjusted for all the others? Please give details.

Discussion and conclusions

1. I’m not sure the reliability and robustness of the analysis warrants the conclusions set out here. I think with further exploration and thinking through of potential selection biases and reasons for missing data, as well as analyses adjusted for appropriate confounders, the results should be reassessed to assure they still support the conclusions. Additionally, greater interpretations of what shared-decision making means to patients and the type of engagement they would want should be included before making conclusions or recommendations about provider engagement.

Is the work clearly and accurately presented and does it cite the current literature?

Yes

If applicable, is the statistical analysis and its interpretation appropriate?

Partly

Are all the source data underlying the results available to ensure full reproducibility?

Yes

Is the study design appropriate and is the work technically sound?

Yes

Are the conclusions drawn adequately supported by the results?

Partly

Are sufficient details of methods and analysis provided to allow replication by others?

Partly

Reviewer Expertise:

Contraceptive side-effects and unmet need, anthropology, demography

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

References : Targets and technologies: Sayana Press and Jadelle in contemporary population policies. Gender, Place & Culture .2020;27(3) : 10.1080/0966369X.2018.1555145 351-369 10.1080/0966369X.2018.1555145 : Fertility regulation as identity maintenance: Understanding the social aspects of birth control. J Health Psychol .2018;23(2) : 10.1177/1359105317726367 240-251 28925281 10.1177/1359105317726367 : Incorporating Method Dissatisfaction into Unmet Need for Contraception: Implications for Measurement and Impact. Stud Fam Plann .2021;52(1) : 10.1111/sifp.12146 95-102 33595116 10.1111/sifp.12146 : Contraceptive Autonomy: Conceptions and Measurement of a Novel Family Planning Indicator. Stud Fam Plann .2020;51(2) : 10.1111/sifp.12114 161-176 32358789 10.1111/sifp.12114 : Understanding Fertility Regulation Strategies among Educated Women in Accra. https://www.popcouncil.org/uploads/pdfs/2016STEPUP_FertilityRegulationAccra.pdf .2016;