The Impact of Telelactation on Breastfeeding Satisfaction at 6 Months Postpartum: Evidence From a Randomized Controlled Trial

Abstract

Objective:

This study evaluates the impact of telelactation (video breastfeeding support visits) on breastfeeding satisfaction and the likelihood of breastfeeding another child. Breastfeeding satisfaction is a key person-centered outcome and indicator of breastfeeding success and is associated with a longer duration of breastfeeding.

Methods:

This randomized controlled trial randomized pregnant individuals to receive a telelactation app (intervention group) or an infant care e-book (control group). The main outcome measured was self-reported breastfeeding satisfaction at 24 weeks postpartum, while a secondary outcome assessed the likelihood of breastfeeding another child. We estimated unadjusted and adjusted linear regression models for the effect of telelactation on breastfeeding satisfaction and logistic regression models for the effect of telelactation on the likelihood of breastfeeding another child. We also examined whether the effects of telelactation differed by breastfeeding problems experienced.

Results:

Results indicated that telelactation significantly improved breastfeeding satisfaction scores by 0.53 points (confidence interval (CI): [0.04,1.04], P = 0.04) and increased the likelihood of participants being very likely to breastfeed another child by 7% (risk ratio = 1.07 [CI: {1.01, 1.14}, P = 0.02]; adjusted risk ratio = 1.07 [CI: {1.01, 1.14}, P = 0.03]). We found no significant differences in outcomes based on race or ethnicity. Further analysis highlighted that telelactation was particularly beneficial for participants experiencing common newborn/premature feeding issues.

Conclusions:

Telelactation can enhance breastfeeding experiences and satisfaction, with implications for public health strategies targeting new parents.

Breastfeeding offers many benefits for infants and birthing people.^1–3 For example, breastfed infants have a lower risk of asthma, obesity, type 1 diabetes, and ear infections.³ However, by 24 weeks postpartum, only 24.9% of infants are breastfed exclusively (ie, not receiving formula or supplementary foods) as recommended by the American Academy of Pediatrics.^4,5 Although public health quality measures have historically focused on breastfeeding rates (eg, duration and exclusivity), breastfeeding satisfaction is also a key, person-centered outcome and indicator of breastfeeding success. Unsurprisingly, several studies have found that satisfaction with breastfeeding is associated with longer duration breastfeeding^6,7 as well as increased likelihood of breastfeeding another child.^8,9 Lewallen¹⁰ found that satisfaction with breastfeeding as measured at 1 week was positively correlated with level of breastfeeding/breastfeeding intensity at 8 weeks postpartum. Further, breastfeeding satisfaction plays a role in the integration of breastfeeding into maternal identity and promotes maternal-infant bonding and maternal well-being.^11–13

Telelactation, video visits between breastfeeding parents and remotely located International Board Certified Lactation Consultants, may play a role in breastfeeding satisfaction. Telelactation can improve access to and the timeliness of professional breastfeeding support.¹⁴ Since 2020, telelactation services have become widely available.¹⁵ Although prior research has shown that telelactation is acceptable and feasible for diverse populations and may improve breastfeeding rates,^16,17 there is minimal evidence on the impact of telelactation on breastfeeding satisfaction.

Professional breastfeeding support through telelactation has the potential to improve breastfeeding self-efficacy and help parents in addressing breastfeeding challenges that may decrease satisfaction. However, remote support may also be insufficient or introduce its own challenges, which could negatively impact breastfeeding satisfaction (eg, technical difficulties in connecting to the call, frustration when certain issues require in-person support). To address these gaps in the literature, we conducted a digital randomized controlled trial (RCT). We found some evidence that telelactation increased breastfeeding rates at 24 weeks, particularly for Black parents.¹⁸ In this study, we aimed to assess the impact of telelactation on secondary trial outcomes, including satisfaction with breastfeeding and the likelihood of breastfeeding another child.

Methods

Data

The Tele-MILC trial was a digital, parallel-design RCT designed to estimate the effect of telelactation support on breastfeeding duration and exclusivity (primary outcomes) and satisfaction (secondary outcome) and was powered to explore treatment effects by race and ethnicity. The power calculation that determined the sample size considered the outcome of any breastfeeding (yes/no) at 6 months postpartum. Complete details on the study design were previously published.¹⁹ Results on the primary outcomes of breastfeeding duration and exclusivity are reported in a separate manuscript.¹⁸ The study was approved by the author’s organizational Institutional Review Board and was registered with ClinicalTrials.gov: NCT04856163.

Pregnant individuals were recruited via advertisements on popular pregnancy apps (Ovia,²⁰ BabyCenter,²¹ What to Expect²²) from July 2021 to December 2022 and were eligible to participate if they were at least 18 years old, pregnant with their first child, in their third trimester of pregnancy, intended to breastfeed, and spoke English and/or Spanish. Exclusion criteria included as follows: non-singleton pregnancy, planned infant separation at birth, advised by a health care provider not to breastfeed, and police custody or incarceration. Individuals who clicked on the study ad were directed to complete an online eligibility screening survey, and eligible respondents then participated in an eConsent process. Those who consented were randomized to receive a telelactation app (treatment), which provided unlimited, on-demand telelactation visits with International Board Certified Lactation Consultants via the Pacify app.²³ Participants randomized to the control group received an e-book on infant care. Participants were surveyed at baseline (the time of enrollment), 4 weeks after their due date, and 24 weeks after delivery date (eg, 24 weeks postpartum).

Outcomes

The main outcome measure was self-reported satisfaction with breastfeeding, as measured at 24 weeks postpartum via a 5-item maternal and infant breastfeeding satisfaction scale, which is a subscale of the validated H&H lactation scale.^24,25 Both the full scale and the 3 subscales have been validated in prior work.¹⁰ Participants were prompted to consider their overall experience with breastfeeding since their baby was born and rate the extent of agreement with 5 statements. Participants who were not currently breastfeeding at the time of the survey (31%) were asked to reflect on their experience when they were breastfeeding. Each item was measured on a 1 to 5 scale, with higher values indicating more agreement with the measures of satisfaction, resulting in a composite satisfaction score ranging from 5 to 25 (Table 1). We note that, although the H&H lactation scale used a 7-point Likert scale for each item, we chose to use a 5-point scale to avoid respondent fatigue and limit the time to complete the survey. This measure was not collected from participants who never initiated breastfeeding.

Table 1.

Breastfeeding Satisfaction Items From Subscale of H&H Lactation Scale

Item
In general, I believe my baby is satisfied with breastfeeding.
In general, I am satisfied with breastfeeding.
I become more relaxed as I sit and breastfeed.
My baby appears to enjoy breastfeeding.
In general, I feel successful at breastfeeding my baby.

Response options include strongly agree, agree, neither agree nor disagree, disagree, and strongly disagree

A secondary outcome in this analysis was self-reported likelihood of breastfeeding another child, as measured at 24 weeks postpartum, ranging from “Very Unlikely” to “Very Likely” on a 5-point scale. We created a dichotomous measure equal to 1 if the participant was “Very likely” to breastfeed another child, and zero, otherwise. This outcome is an important prospective measure of future breastfeeding intention, which is closely related to satisfaction and experience with breastfeeding and may be less subject to interpretative challenges than questions directly assessing satisfaction.

Breastfeeding Problems

The 24-week survey asked respondents who were still breastfeeding to select all infant-related breastfeeding challenges they experienced from birth to present from the following list: latching, choking, not waking up to breastfeed often enough, not being interested in breastfeeding, getting distracted, wanting to breastfeed too often, and problems gaining weight. Respondents were also asked about problems related to the parent’s breasts and milk supply, including not having enough milk, taking too long for milk to come in, trouble getting milk flow to start, sore nipples, engorgement, infections, clogged ducts, and leaking. Respondents who were no longer breastfeeding at 24 weeks were not asked to report problems they experienced over the full course of breastfeeding, as the survey included skip patterns to limit the survey length to 15 minutes.

Statistical Methods

We estimated unadjusted and adjusted linear regression models for the effect of telelactation on breastfeeding satisfaction and binomial generalized linear regression models (with a log link function) for the effect of telelactation on the likelihood of breastfeeding another child. Adjusted analyses were performed to adjust for chance imbalance of baseline characteristics²⁶ and included select prognostic covariates previously hypothesized to be associated with breastfeeding outcomes, including maternal education, maternal race and ethnicity, health insurance coverage, presence of a chronic condition, and weeks of gestation at delivery. Item nonresponse was present for weeks of gestation and addressed using the missing indicator method.²⁷

Because our study was powered to detect differences in primary outcomes by race and ethnicity, we also assessed heterogeneity of treatment effects by race and ethnicity by interacting the treatment dummy variable with maternal race and ethnicity categorized into 3 mutually exclusive categories: Black, Latinx, and non-Black and non-Latinx. We interpret the P-value for the joint test of significance of the interaction term to determine whether there is evidence for heterogeneous effects.

We conducted intention-to-treat analyses including all study participants with nonmissing 24-week survey outcome data. For analyses of the effect of telelactation on outcomes, a sensitivity analysis used inverse probability weighting (IPW) to adjust for missing outcome data, following guidance for missing outcomes in RCTs.²⁸ Specifically, we used logistic regression to estimate the probability of response for all participants who were randomized, including as covariates all variables used in adjusted models as well as additional baseline variables associated with breastfeeding outcomes in the literature (cesarean section birth mode, maternal [or household] income, plans to return to work in the first year). Next, we used the inverse of the predicted probability of response as weights in our regression models.²⁹ These weighted regression models were estimated based on the same subset of individuals with 24-week survey outcome data as for our main analysis.

We hypothesized that the effect of telelactation could differ depending on the types of problems experienced. For example, certain problems (eg, latching) may be more amenable to resolution through real-time support, in which case telelactation may offer significant benefits. Other problems might be difficult for International Board Certified Lactation Consultants to address virtually or at all (eg, mastitis/infections that may require treatment by a local health care provider who can prescribe medication). We therefore conducted exploratory analyses related to breastfeeding problems during the first 24 weeks postpartum, as reported by parents who were still breastfeeding when they responded to the 24-week survey. Because there were so many different types of problems respondents could select, and many are related, we grouped problems using factor analysis, with the number of factors determined using parallel analysis.³⁰ We then assigned each problem to the factor with the maximum loading, and defined new composite measures of problems as the total number of problems in each factor. We then fit unadjusted and adjusted regression models for breastfeeding satisfaction and likelihood of breastfeeding another child, as described above, but including these composite measures of problems as covariates.

Analyses were completed in September to October of 2024 and performed using R version 4.2.2.

Results

A total of 2108 participants were enrolled in the trial (n = 1052 intervention (telelactation) arm; n = 1056 control arm), and 1908 participants (n = 952 telelactation; n = 956 control) completed the final assessment and responded to the 24-week survey outcomes. Baseline characteristics were similar for telelactation and control arm participants (Table 2). For the breastfeeding satisfaction measure, 59 individuals never initiated breastfeeding (and thus satisfaction was not assessed), leaving 1849 participants included in the analysis of breastfeeding satisfaction.

Table 2.

Self-reported Baseline Characteristics of the Sample

	Control (n = 956) n (%) or Mean (SD)	Telelactation (n = 955) n (%) or Mean (SD)	P-Value
Race and ethnicity
Black	310 (32.4)	301 (31.5)	0.91
Latino	336 (35.1)	342 (35.8)
White	310 (32.4)	312 (32.7)
Private health insurance	602 (0.63)	611 (0.64)	0.83
Comorbidities	430 (0.45)	420 (0.44)	0.76
Education			0.30
High school or less	122 (12.8)	124 (13.0)
Some college	287 (30.0)	269 (28.2)
4-year college	275 (28.8)	313 (32.8)
Graduate degree	272 (28.5)	249 (26.1)
Married	553 (57.8)	589 (61.7)	0.10
Gestational age	38.9 (1.4)	39.0 (1.3)	0.31
C-section	293 (30.6)	280 (29.3)	0.56

Breastfeeding satisfaction scores were 0.53 points higher (confidence interval (CI): [0.04, 1.04], P = 0.04) in the telelactation arm compared to the control arm (telelactation: M = 19.49, SD = 5.43; control: M = 18.95, SD = 5.60). Adjusted linear models resulted in a similar estimated effect of telelactation of 0.49 (CI: [−0.01, 0.99], P = 0.05). Sensitivity IPW analyses found slightly larger effects of telelactation (Table 3).

Table 3.

Treatment Effects for Breastfeeding Satisfaction and Likelihood of Breastfeeding Another Child

Outcomes	Mean or %		ITT-Complete Case		ITT-IPW
	Control	Telelactation	Unadjusted Difference or RR (95% CI), P	Adjusted* Estimate or RR (95% CI), P	Unadjusted Estimate or RR (95% CI), P	Adjusted* Estimate or RR (95% CI), P
Breastfeeding satisfaction	19.0	19.5	0.53 (0.04, 1.04), P = 0.04	0.49 (−0.01, 0.99), P = 0.05	0.66 (0.15, 1.16), P = 0.01	0.64 (0.14, 1.14), P = 0.01
Very likely to breastfeed another child	65%	70%	1.07 (1.01, 1.14), P = 0.02	1.07 (1.01, 1.14), P = 0.03	1.08 (1.02, 1.15), P = 0.01	1.08 (1.02, 1.15), P = 0.01

IPW indicates inverse probability weighting; ITT, intention-to-treat.

Risk ratios (RRs) are presented for the binary outcome “Very likely to breastfeed another child.”

^*Adjusted analyses included the following covariates: maternal education (4 categories), maternal race and ethnicity, an indicator for private health insurance during pregnancy, an indicator for whether the participant reported a chronic condition, and an indicator for delivery at 39 or greater weeks of gestation.

The risk ratio for being very likely to breastfeed another child was 1.07, meaning the probability of being very likely to breastfeed was 7% larger among telelactation participants relative to the control arm (RR = 1.07 (CI: [1.01, 1.14], P = 0.02; adjusted risk ratio = 1.07 (CI: [1.01, 1.14], P = 0.03). Sensitivity IPW analyses were similar (Table 3). There was no evidence of heterogeneity of telelactation effects by race and ethnicity for either the breastfeeding satisfaction or likelihood to breastfeed another child measure.

Factor analysis based on problems reported on the 24-week survey for 1311 respondents still breastfeeding at 24 weeks resulted in 5 domains (Table 4): 1) undersupply (milk took too long to come in, trouble getting the flow to start, not enough milk), 2) baby-centered feeding issues (baby choked, was distracted, wanted to nurse too often), 3) breast pain/health issues (sore nipples, clogged ducts, infection), 4) common newborn/premature feeding issues (baby had trouble latching, would not wake up, not interested, did not gain enough weight), and 5) breast engorgement/leakage (oversupply domain). These domains, suggested by factor analysis, are sensible, as they tend to group related problems together into domains (Table 4). Telelactation appeared to have a larger effect on breastfeeding satisfaction for those with “common newborn/premature feeding issues.” For each additional problem in this domain (containing 4 problems), the effect of telelactation on satisfaction increased by 0.51 in unadjusted analyses (P = 0.03) and 0.47 in adjusted analyses (P = 0.04). Thus, telelactation would be expected to improve satisfaction by approximately 2 points more for someone experiencing all 4 of these problems compared to someone without any of these problems. There was no evidence of differential effects of telelactation for any other domains. There was no evidence of differential effects of telelactation on the likelihood of breastfeeding another child by breastfeeding problems in either unadjusted or adjusted analyses.

Table 4.

Groupings of Problems From Factor Analysis

Problems	Domain
Milk took too long to come in, trouble getting flow to start, not enough milk	Undersupply
Baby choked, was distracted, wanted to nurse too often	Baby-centered feeding issues
Sore nipples, clogged ducts, infection	Breast issues
Baby had trouble latching, would not wake up, not interested, did not gain enough weight	Common newborn/premature feeding issues
Breasts leaked, breasts were engorged	Oversupply

Discussion

Results showed that access to telelactation resulted in statistically significant improvements in breastfeeding satisfaction and the likelihood of breastfeeding another child. Further, telelactation appeared to have a larger effect on breastfeeding satisfaction among birthing individuals experiencing particular types of breastfeeding problems, such as common newborn/premature feeding issues. This study is the first RCT that was powered to assess the impact of telelactation video visits on breastfeeding outcomes, including breastfeeding satisfaction, and results suggest this form of professional breastfeeding support can improve the breastfeeding experiences of new parents and increase the likelihood of breastfeeding success. It is promising that we did not observe heterogeneous treatment effects by race and ethnicity. This suggests that implementation of a telelactation service is not likely to increase disparities in breastfeeding.

While a difference of 0.53 points in breastfeeding satisfaction across study arms may seem like a small effect size, it should be considered in light of several factors. First, satisfaction scores were generally high overall (mean of 19 points on a 5–25 point scale).^31,32 Second, we observed a statistically significant effect of telelactation at 6 months postpartum despite the fact that we have demonstrated in other work that the majority of telelactation visits occur in the immediate postpartum period.³³ Thus, one might expect differences in breastfeeding satisfaction due to telelactation to be more evident in the early postpartum period. Improvements at 6 months postpartum suggest that early telelactation support can have lasting impacts on breastfeeding satisfaction, and by extension, breastfeeding behaviors. Finally, most studies of breastfeeding support interventions find minimal or no effects on breastfeeding satisfaction as measured using the H&H or MBFES lactation scales.^31,32 For example, a pragmatic trial assessing a Baby-Friendly Initiative observed similarly high breastfeeding satisfaction scores in general, and detected no significant differences across intervention and control arms.³²

It is especially noteworthy that telelactation increased the odds of being very likely to breastfeed another child by 25%. This finding provides evidence of telelactation’s effectiveness but can also inform program design and implementation decisions. For example, targeting telelactation to first-time parents may influence behaviors with subsequent children, increasing the broader public health impact of this type of support.

Limitations

There are several limitations. First, given limited research using the maternal and infant breastfeeding satisfaction subscale of the H&H lactation scale, we are not able to comment on whether the improvements observed here are clinically meaningful. Further, the scale has primarily been used among individuals who were breastfeeding at the time of assessment, so application to those who have stopped breastfeeding (the minority of individuals in the study) is relatively new and requires further validation. Second, survey responses can be affected by recall bias. For example, some participants may have had difficulty recalling all of the breastfeeding problems they experienced, especially in the early weeks of breastfeeding. Further, the subset of participants who were no longer breastfeeding at the time of the 24-week survey were asked to reflect on their satisfaction when they were breastfeeding. Third, participants were recruited through pregnancy apps and demonstrated existing engagement with mobile health applications, potentially representing a demographic distinct from the broader US population. Fourth, survey questions on breastfeeding problems were detailed, and responses may be affected by the limited attention of respondents. In addition, only participants who reported any breastfeeding at 24 weeks postpartum (the majority of participants, but not all) received the breastfeeding problems survey questions. Finally, breastfeeding satisfaction varies over time, and our study only assessed satisfaction at 24 weeks postpartum, a period when breastfeeding is well-established and problems are less common.

Conclusion

This study provides evidence of the effectiveness of telelactation in enhancing maternal breastfeeding satisfaction and the likelihood of breastfeeding subsequent children. Despite its infrequent use as a person-reported outcome measure, maternal satisfaction is a key outcome, and our findings highlight the importance of supports that prioritize the experiences and needs of breastfeeding parents.

The results point to telelactation as a valuable resource for diverse populations. Additionally, the identification of specific breastfeeding challenges that telelactation can effectively address provides insights for tailoring lactation support services. Future research should explore the effectiveness of different virtual and in-person breastfeeding support models, with consideration of a wider range of person-centered outcomes.

WHAT’S NEW.

We examined telelactation’s impact on breastfeeding satisfaction and future breastfeeding intentions. Results show telelactation increased satisfaction and likelihood to breastfeed again, with no evidence of differential effects across racial groups. Telelactation is promising for improving breastfeeding satisfaction in diverse populations.