This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Research Protocols, is properly cited. The complete bibliographic information, a link to the original publication on https://www.researchprotocols.org, as well as this copyright and license information must be included.
Risky drinking is prevalent among women of childbearing age. Although many women reduce their drinking during pregnancy, more than half return to prepregnancy levels during the early postpartum period. Risky drinking in new mothers may be associated with negative child and maternal health outcomes; however, new mothers are unlikely to seek treatment for risky drinking because of stigma and fear of child protective service involvement. SMS text messaging is a promising approach for reaching non–treatment-seeking new mothers at risk because of risky drinking. SMS text messaging interventions (TMIs) are empirically supported for alcohol use, but a tailored intervention for new mothers does not exist. This study aims to fill this gap by developing a just-in-time adaptive TMI for postpartum risky drinking.
The objectives of this paper are to present a preliminary conceptual model of postpartum risky drinking and describe the protocol for conducting an ecological momentary assessment (EMA) study with new mothers to inform the refinement of the conceptual model and development of the TMI.
This paper presents a preliminary conceptual model of postpartum risky drinking based on the motivational model of alcohol use, social cognitive theory, and temporal self-regulation theory. The model proposes three primary intervention targets: motivation, self-efficacy, and self-regulation. Theoretical and empirical literature in support of the conceptual model is described. The paper also describes procedures for a study that will collect EMA data from 30 participants recruited via social media and the perinatal Central Intake system of New Jersey. Following the baseline assessment, EMA surveys will be sent 5 times per day for 14 days. The assessment instruments and data analysis procedures are described.
Recruitment is scheduled to begin in January 2022 and is anticipated to conclude in March 2022. Study results are estimated to be published in July 2022.
The study findings will enhance our understanding of daily and momentary fluctuations in risk and protective factors for risky drinking during the early postpartum period. The findings will be used to refine the conceptual model and inform the development of the TMI. The next steps for this work include the development of intervention components via an iterative participatory design process and testing of the resulting intervention in a pilot microrandomized trial.
PRR1-10.2196/36849
Risky drinking, defined as the consumption of ≥4 drinks in a day or ≥8 drinks in a week [
SMS text messaging is a promising approach for reaching non–treatment-seeking risky drinkers and may be particularly suitable for addressing postpartum risky drinking. With 97% of Americans owning a cell phone as of 2021, access to SMS text messaging is widespread [
A growing body of literature demonstrates support for TMIs in reducing risky drinking in non–treatment-seeking adults [
TMIs that tailor content to specific participant characteristics or clinical needs show larger effects on clinical outcomes and lower rates of attrition than programs that deliver generic messages [
This study will apply the multiphase optimization strategy (MOST) framework [
Following the MOST framework, the first step in developing a theory-driven JITAI for postpartum risky drinking is to specify the theoretical pathways from the postpartum risk factors that are the primary intervention targets to the ultimate desired outcome of reduced risky drinking. Our proposed conceptual model (described in the
As JITAIs aim to intervene at the momentary level, a comprehensive understanding of the daily and momentary fluctuations in risks and protective factors for postpartum risky drinking is needed to inform the design of a tailored JITAI for this population. EMAs collect data in real time over the course of a day and are designed to capture momentary fluctuations in feelings and behaviors as participants go about their daily lives [
The objectives of this paper are to (1) present a preliminary conceptual model of postpartum risky drinking and (2) describe the protocol for conducting an EMA study with new mothers to inform the refinement of the conceptual model and the development of a just-in-time adaptive TMI to address postpartum risky drinking.
The purpose of the EMA study is to assess and refine the conceptual model and test the feasibility of EMA data collection procedures in a sample of new mothers. The primary research questions of the EMA study are as follows: (1) How do momentary and daily fluctuations in internal and external contextual factors affect motivation, self-efficacy, and self-regulation? (2) Which internal and external factors are most salient at particular times of the day? (3) What is the relationship between maternal self-efficacy and drinking self-efficacy and how does this relationship fluctuate throughout the day? (4) What is the relationship between momentary and daily changes in motivation, self-efficacy, and self-regulation and daily drinking? (5) To what extent are the study methods (eg, number and length of surveys and item wording) acceptable and feasible for the target population of new mothers within the early postpartum period?
This EMA study is part of a 3-year effort to develop a JITAI for postpartum risky drinking that comprises the preparation phase of the MOST framework (
Study timeline. EMA: ecological momentary assessment; JITAI: just-in-time adaptive intervention; MRT: microrandomized trial.
The study was approved by the Solutions Institutional Review Board in October 2020 (#2020/06/15) and registered at ClinicalTrials.gov. All study participants will provide informed consent to take part.
Our preliminary conceptual model was developed via a review of relevant theoretical and empirical literature combined with a series of brainstorming conversations among the study team.
Preliminary conceptual model of postpartum risky drinking. SE: self-efficacy.
According to the motivational model of alcohol use, motivation to drink is the most proximal predictor of drinking behavior [
The motivational model proposes four types of drinking motives that may affect within-day fluctuations in motivation to drink: coping (aimed at reducing negative emotions), enhancement (aimed at increasing positive emotions), conformity (aimed at avoiding social rejection), and social (aimed at increasing positive social experiences), with varying antecedents and consequences of each [
Consistent with social cognitive theory, the conceptual model suggests that two types of self-efficacy—drinking self-efficacy and maternal self-efficacy—may contribute to drinking behavior. Drinking self-efficacy, defined as a person’s belief in their ability to avoid drinking, is well-supported as a significant predictor of drinking behavior. Higher drinking self-efficacy has been shown to predict less drinking and improved long-term outcomes in the context of treatment for alcohol use disorder [
In addition to drinking self-efficacy, our conceptual model includes self-efficacy specific to the maternal role, defined as a mother’s belief in her ability to successfully care for her baby. There is currently no research examining associations between maternal self-efficacy and alcohol use. In addition, no study to date has examined daily or momentary changes in maternal self-efficacy during the postpartum period despite evidence that self-efficacy in other domains changes over brief periods [
Both drinking self-efficacy and maternal self-efficacy are influenced by internal and external contextual factors, as reflected in the conceptual model. Variations in mood, stress, and fatigue have been shown to affect both drinking self-efficacy [
According to temporal self-regulation theory, self-regulation, or the ability to monitor and adapt cognitions, emotions, and behaviors in response to internal or external contextual factors in a goal-directed manner, is a key factor affecting risky behaviors, including alcohol use [
A small number of studies have examined daily within-person changes in self-regulation strategies in the context of alcohol use [
The study target population is adults aged 18-45 years who live in New Jersey and gave birth to a live infant within the previous 2 weeks who is currently in their care. This study is being conducted in New Jersey to leverage existing partnerships between the study team and the New Jersey state system of care for perinatal women. Additional eligibility criteria include speaking English and access to a smartphone with internet. Participants must also report one of the following: (1) a score of ≥2 on the Tolerance, Annoyance, Cut Down, Eye-Opener (T-ACE) alcohol risk screener, (2) having ≥8 standard drinks in 1 week in the 12 months before becoming pregnant, or (3) having ≥4 drinks at one time once a month or more often in the 12 months before becoming pregnant. We aim to recruit 30 participants who meet the eligibility criteria. Similar sample sizes have been used in other EMA studies of individuals who engage in substance use [
This study will use two primary recruitment strategies: (1) recruitment via social media advertisements on Facebook and Instagram and (2) referrals from providers in the New Jersey perinatal Central Intake (CI) system.
Advertisements for the study will be placed on Facebook and Instagram and will be geographically targeted to New Jersey. Additional advertisement targeting will include interests related to birth, pregnancy, motherhood, infant care, and drinking alcohol. Individuals who click on an advertisement will be directed to the study website. Social media recruitment via Facebook and Instagram is widely used in research study recruitment and has been used successfully with both new mothers [
New Jersey operates a state-wide CI system that provides a single point of entry into services for pregnant and postpartum people to promote improved care coordination and access to needed services. For this study, we will partner with one of the CI sites that serves a large, demographically diverse county in the state. CI workers will introduce the study to their clients using a script provided by the study team. If a client is interested in learning more about the study, the CI worker will provide their contact information to the study coordinator, who will contact the client within 2 weeks of her due date to complete eligibility screening. For clients who are interested in the study but prefer not to share their contact information, the CI worker will direct them to the study website, where they can complete eligibility screening and informed consent on the web.
Eligible participants who complete the informed consent process will be invited to complete the baseline survey. Baseline survey data will be used to describe the study sample and understand the impact of baseline characteristics on momentary changes in the variables of interest. The baseline survey can be completed either on the web via Qualtrics (Qualtrics International Inc) or via phone with the study coordinator, depending on the participant’s preference. The baseline survey will take approximately 30 minutes to complete, and participants will receive a US $25 gift card upon completion. The baseline survey will assess demographic characteristics, maternal self-efficacy, mental health, stress and coping, motivation, alcohol use, and other substance use. See
Baseline measures.
Construct | Description | Measure (reference) |
Demographics | Age, gender, marital status, race, ethnicity, living arrangements, childbirth history, education, employment, income, and substance use treatment history | —a |
Drinking self-efficacy | Perceived ability to handle various drinking situations | Drinking Refusal Self-efficacy Questionnaire–Revised [ |
Alcohol and drug use history | Use of alcohol, marijuana, and illegal drugs before pregnancy, during pregnancy, and since giving birth | Adapted from the NIDAb-modified ASSISTc [ |
Motivation | Readiness to change alcohol use | Maternal Motivation Scale [ |
Postpartum depression | Symptoms of depression since giving birth | Beck Depression Scale [ |
Maternal self-efficacy | Confidence in carrying out various baby care tasks | Karitane Parenting Confidence Scale [ |
Trauma history | Experiences of trauma during childhood | Adverse Childhood Experiences Questionnaire [ |
Pandemic stress | Stress related to the COVID-19 pandemic | Adapted from the Pandemic Stress Index [ |
Attachment to infant | Mother experience of bonding and attachment to baby | Infant Bonding Scale [ |
Fatigue | Experiences of emotional and physical fatigue | Fatigue Assessment Scale [ |
Stress | Perceptions of stress related to general life experiences | Perceived Stress Scale–4 [ |
Drinking motives | Motivation to consume alcohol | Drinking Motives Questionnaire–Revised [ |
Coping self-efficacy | Perceived ability to cope with challenging life events | Coping Self-efficacy Scale [ |
Social support | Perceptions of social support | Interpersonal Support Evaluation List–12 [ |
Digital literacy | Comfort using technology to complete tasks, such as SMS text messaging, using a smartphone, and accessing health information on the web | Media and Technology Usage and Attitudes Scale [ |
aThere is no specific citation for the demographic items.
bNIDA: National Institute on Drug Abuse.
cASSIST: Alcohol, Smoking, and Substance Involvement Screening Test.
dNIAAA: National Institute on Alcohol Abuse and Alcoholism.
All EMA data will be collected via the MetricWire app. The MetricWire app is available for free download from the Apple App Store and Google Play Store and has been used in other EMA research studies [
Sample schedule of daily survey prompts.
The remainder of the day will be divided into 4 equal segments, and hourly surveys will be sent randomly within each segment. No surveys will be sent later than 9 PM. After each survey prompt, the survey will be available for up to 60 minutes, with 2 reminder prompts sent at 20 and 40 minutes. Surveys that are not completed within the 60-minute window will expire. Completion of each hourly survey will take 1-2 minutes.
Between the last survey in the evening and the first survey in the morning, an optional EMA survey will be available for the participants to complete. The reason for this optional survey is to enable data collection during the night, when the participants may be awake with their baby. Nights may be times of high stress and high risk of drinking for new mothers. This survey will allow us to capture data on these middle-of-the-night times without disturbing the participants by sending prompts. The participants who complete the night survey will receive an automatic response SMS text message with contact information for a 24-hour support hotline.
Study participants will be remunerated for taking part in the study in the form of gift cards to Amazon or Target. The participants will be paid US $25 for completion of the baseline survey. During EMA data collection, the participants will be paid US $2 for each EMA survey completed, with a bonus of US $20 for completing >50% of the EMA surveys and US $30 for completing >80% of the EMA surveys. Bonus incentives are used routinely in EMA studies to boost compliance and have been used in EMA studies with postpartum women [
To ensure that the participants are adequately supported during the study, we will engage in the following: (1) check in briefly by phone with all participants after 3 days of EMA to obtain initial feedback on the questions and address any technical difficulties, (2) provide information on how to obtain immediate support via hotlines, and (3) provide all participants with a list of local mental health and substance use treatment and support resources at the outset of the study. Information about how to obtain immediate support will be available within the MetricWire app at all times for the participants to access as needed.
The study team reviewed the existing literature and selected EMA measures that align with each construct in the conceptual model (
Morning and hourly ecological momentary assessment (EMA) survey items and response options.
Construct, subcategories, and measure (reference) | Item (response options) | Morning | Hourly | |||||
|
||||||||
|
|
|||||||
|
|
Adapted from the studies by Nguyen et al [ |
What is your overall feeling right now? (1=very unpleasant to 7=very pleasant) What is your overall energy level right now? (1=very low to 7=very high) What is your anxiety level right now? (1=very low to 7=very high) |
|
✓ | |||
|
|
|||||||
|
|
Adapted from the Perceived Stress Scale [ |
What is your overall stress level right now? (1=very low to 7=very high) |
|
✓ | |||
|
|
|||||||
|
|
Adapted from the Fatigue Assessment Scale [ |
How physically exhausted are you right now? (1=not at all to 7=extremely) How mentally exhausted are you right now? (1=not at all to 7=extremely) |
|
✓ | |||
|
|
Adapted from the studies by Dennis and Ross [ |
How well did your baby sleep last night? (1=poor to 7=excellent) How well did you sleep last night? (1=poor to 7=excellent) |
✓ |
|
|||
|
||||||||
|
|
|||||||
|
|
Selected items from the Infant Characteristics Questionnaire [ |
Yesterday, how easy or difficult was it for you to calm or soothe your baby when they were upset? (1=very easy to 7=very difficult) Yesterday, how much did your baby cry and fuss in general? (1=very little to 7=a lot) |
✓ |
|
|||
|
|
Adapted from the study by Adams et al [ |
Since the last survey, how many times did your baby fuss, cry, or seem upset? (0, 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 or more times) How many of these times were you able to successfully soothe your baby? (All of them; most of them; some of them; a few of them; none of them) |
|
✓ | |||
|
|
|||||||
|
|
Item from the Maternal Social Support Index [ |
How often was support available to you when you needed it yesterday? (None of the time; little of the time; some of the time; most of the time) |
✓ |
|
|||
|
|
|||||||
|
|
Adapted from the study by McQuoid et al [ |
Was alcohol available to you yesterday? (Yes or no) Were other people in your household drinking alcohol yesterday? (Yes or no) |
✓ |
|
|||
|
||||||||
|
|
|||||||
|
|
Selected items from the Drinking Motives Questionnaire [ |
If you drank alcohol yesterday, why did you drink? (Because it makes social gatherings more fun; to forget about your problems; because it gives you a pleasant feeling; to be liked) |
✓ |
|
|||
|
|
|||||||
|
|
Adapted from the study by Kuerbis et al [ |
How committed are you to not drink alcohol in the next hour (for hourly survey) or today (for morning survey)? (1=not at all to 7=extremely) |
✓ | ✓ | |||
|
||||||||
|
|
|||||||
|
|
Developed for this study |
How confident are you that you will be able to meet your baby’s physical needs (such as needs to be fed or changed) over the next hour (for hourly survey) or today (for morning survey)? (1=not at all confident to 7=extremely confident) How confident are you that you will be able to meet your baby’s emotional needs (such as needs to be soothed or entertained) over the next hour (for hourly survey) or today (for morning survey)? (1=not at all confident to 7=extremely confident) |
✓ | ✓ | |||
|
|
|||||||
|
|
Adapted from the study by Kuerbis et al [ |
How confident are you that you can avoid drinking alcohol for the next hour (for hourly survey) or today (for morning survey)? (1=not at all confident to 7=extremely confident) |
✓ | ✓ | |||
|
||||||||
|
|
|||||||
|
|
Adapted from the studies by Roos et al [ |
Did you use any strategies to cope with negative feelings or stress since the last survey (for hourly survey) or yesterday (for morning survey)? (I didn’t experience negative feelings or stress; I drank alcohol; I changed my thinking; I changed my current situation; I found something else to do; I sought advice or support; I came up with a plan to cope; I set a goal or kept track of my current progress toward a goal; I directly communicated my needs to others; I tried to relax; I took medication; I pushed negative feelings or stress away; I used another strategy; I didn’t use any strategies) |
✓ | ✓ | |||
|
|
|||||||
|
|
Adapted from the studies by Roos et al [ |
Did you use any strategies to manage the urge to drink alcohol since the last survey (for hourly survey) or yesterday (for morning survey)? (I didn’t experience an urge to drink alcohol; I changed my thinking; I changed my current situation; I found something else to do; I sought advice or support; I came up with a plan to manage the urge to drink alcohol; I set a goal or kept track of my current progress toward a goal; I directly communicated my needs to others; I tried to relax; I made an effort to stay safe and avoid risks while drinking alcohol; I took medication; I tried to ignore the urge to drink alcohol; I used another strategy; I didn’t use any strategies) |
|
|
|||
|
||||||||
|
|
|||||||
|
|
Selected items from the Perceived Maternal Parenting Self-efficacy Tool [ |
Yesterday, I was good at feeding my baby (1=strongly disagree, 2=disagree, 3=agree, 4=strongly agree) Yesterday, I was good at soothing my baby when they became upset (1=strongly disagree, 2=disagree, 3=agree, 4=strongly agree) Yesterday, I was good at reading my baby’s cues (1=strongly disagree, 2=disagree, 3=agree, 4=strongly agree) Yesterday, my baby responded well to me (1=strongly disagree, 2=disagree, 3=agree, 4=strongly agree) |
✓ |
|
|||
|
|
|||||||
|
|
Developed for this study |
How strong was your urge to drink alcohol yesterday? (1=very low to 7=very high) To what extent did you feel that you overcame your urge to drink alcohol yesterday? (1=not at all to 7=to a very great extent) |
✓ |
|
|||
|
||||||||
|
|
|||||||
|
|
Adapted from the studies by O’Donnell et al [ |
Did you consume any alcohol yesterday? (Yes or no) If yes, how many alcoholic drinks did you consume? (Enter numeric value) If yes, select all the time periods when you drank alcohol yesterday (Morning; lunchtime; afternoon; evening; during the night) |
✓ |
|
Drawing from other EMA feasibility studies [
We will apply the following established methods to encourage and monitor compliance: (1) one-on-one training on EMA procedures before the start of data collection [
As the primary purpose of the EMA is to inform the development of the JITAI, the analyses will be largely descriptive. To avoid issues of data quality stemming from noncompliance, we will exclude participants who complete <50% of the required EMA surveys [
Recruitment for this study is scheduled to begin in January 2022. We anticipate completing recruitment and enrollment by March 2022 and expect to have completed EMA data collection by April 1, 2022. Study results will be published in peer-reviewed scientific journals upon completion of data analysis, which is estimated to be in July 2022.
This paper presents a preliminary conceptual model of postpartum risky drinking as well as a protocol for an EMA data collection study aimed at refining the conceptual model and informing the development of the first JITAI for postpartum risky drinking. This study is the first to assess in-the-moment predictors of risky drinking in the postpartum period and will thus fill critical gaps in existing research. New mothers who engage in risky drinking and other substance use are understudied and underserved as much of the intervention research on perinatal substance use is focused on pregnancy despite high risks of increasing substance use in the early postpartum weeks [
The study findings will enhance our understanding of daily and momentary fluctuations in risk and protective factors for risky drinking during the early postpartum period, a time when risk for alcohol use is high and access to treatment is often low [
A primary strength of this study is the reliance on theory to guide EMA data collection and JITAI development. A recent systematic review of JITAIs for substance use found that most existing studies did not apply state-of-the-art methods such as the MOST framework and did not sufficiently incorporate theory into intervention development [
The inclusion of variables that are especially salient in the postpartum period, such as maternal self-efficacy, baby irritability, and sleep, is an additional strength of this study. Studies of mothers in substance use treatment demonstrate a complex relationship between motherhood and substance use treatment and recovery. Although motherhood and caring for children are often described as a critical motivating factor for seeking treatment and reducing substance use [
Study limitations include the requirement to speak and read English and own a smartphone with internet access, limiting generalizability. In addition, the study is focused on alcohol use only, which may leave needs related to the use of other substances unaddressed. Many pregnant and postpartum people who engage in risky drinking also use other substances [
The need for tailored digital supportive interventions for the postpartum period is greater than ever given the increasing rates of perinatal stress, depression, and substance use during the COVID-19 pandemic [
Despite the limitations, this study has the potential to significantly contribute to the existing literature by improving our understanding of the antecedents of postpartum risky drinking and informing the development of a tailored JITAI to address it. If feasibility is supported, the EMA protocol can also serve as a model for future studies that aim to collect real-time data from new mothers. This study represents a first step in a larger program of research aimed at using technology to reach underserved new mothers with interventions for perinatal substance use that are evidence-based and tailored to their identities as mothers and aim to empower mothers to seek help while reducing stigma and fear. The methods and findings of this study will be applied to future efforts to ultimately expand the JITAI to include other substances beyond alcohol as well as to create culturally tailored versions. Significant racial and ethnic disparities exist in access to support and treatment for substance use, with Black and Latinx mothers experiencing higher levels of stigma and greater access barriers to obtaining needed support [
National Institute on Alcohol Abuse and Alcoholism grant reviews.
Central Intake
ecological momentary assessment
just-in-time adaptive intervention
multiphase optimization strategy
text messaging intervention
This study is funded by the National Institute on Alcohol Abuse and Alcoholism (grant R34AA028407; see
None declared.