In this study, we interviewed 13 female SOs sharing a child with a male co-parent with UAU after they had participated in a web-based self-delivered support program called SPARE.

We found that the main reasons for seeking support were connected to the SOs’ own emotional needs deriving from co-parent alcohol consumption and trying to understand or cope with consequences of the co-parent’s drinking. These findings are in line with the widely used Stress–Strain-Coping-Support (SSCS) model [35]. The model has been developed after decades of studies on SOs in various cultures. Common themes described by SOs who are seeking support are for example threats to the SOs and the family, difficulties with coping strategies, barriers to receiving support, and increased levels of own mental health problems [4, 8]. The reasons described by the SOs in our study are coherent with the previous research which shows that the SSCS model is applicable to SOs with differing levels of accumulated burdens.

The results are also consistent with those of a qualitative study in Australia analyzing themes for help-seeking in online counselling transcripts of partners contacting a national web-based service for substance and alcohol use [36]. The study found three broad themes with seven sub-themes, including Seeking advice, Wanting to talk, Discussing help-seeking, and Coping processes. The authors concluded that web-based counselling was important in facilitating support-seeking for SOs, since it lowered the barriers and could be a first step to seek further support, if it was insufficient in itself—a conclusion that is supported by the results of our current study. Further, most participants in our study described being reluctant to contact social services. This is in line with the results of a study on barriers and facilitators regarding SO help-seeking by McCann et al. [30], which showed that previous negative experiences of authorities, self-stigma, and public stigma were barriers. We argue that the anonymity provided in SPARE was a key aspect in facilitating for SOs to enter the support program.

The gender asymmetry in our sample reflected the RCT, where 96% of the participants were female, a proportion similar to those in all previous CRAFT trials, with women constituting 72–100% of the participants [28]. Studies from several different cultures confirm that the burden on female SOs is greater than that on male SOs due multiple responsibilities such as taking care of both children and husband or risk of intimate partner violence [8, 37]. Additionally, sharing a child often comes with greater strains on women [11, 38]. A co-parent with UAU adds a burden to the women taking care of the family. We interpreted the SOs’ statements about unreliable, dull, absent, selfish, and mean partners, and the constant low-key stress that at times erupted in major stressful events, as descriptions of such strains. In essence, we suggest that Swedish female SOs also experience heavier burdens than male SOs, explaining at least part of the gender asymmetry.

Regarding perceived effects of the SPARE program, most SOs described feeling calmer and relieved, had started to put their own needs first, and adapted less to the co-parent. This is in line with the results of a study by Hellum et al. [39], where self-delivered CRAFT was one of three forms of program distribution (the others being face-to-face individual sessions and group sessions) and the written material was found to be helpful mainly in improving SOs’ quality of life. Hellum et al. discuss the importance of SOs being met in a non-judgmental and accepting manner to increase awareness of their not being alone and feeling acknowledged as living under stressful circumstances. We draw the same conclusions from the interviews in our study.

Although almost none of the SOs mentioned looking specifically for parenting strategies when searching for support, almost everyone described how one of the main appreciated effects of SPARE was related to an increased focus on their children. After setting aside 15 min daily for dedicated parent–child time, the SOs described fewer conflicts with the children, having more fun together, and talking more openly to each other. We argue that a new insight from this study is how the 15-min strategy can be applied in a situation where the main problem is not children with disruptive behaviors or a malfunctioning parent–child relationship, but where one parent’s shortcomings (in this case UAU) lead to neglect and other forms of negative consequences for the children. Dedicated parent–child time seemed to help the SOs shift focus away from arguments with the co-parent, towards the children. The positive effects on the relationship with the children were clearly described in several interviews. The protective factor of a good relationship with one parent when the other parent has substance use problems is well-established [9, 16, 18] and we suggest that the present study shows that one of the main effects of SPARE is an improved parent–child relationship. Hence, we would propose that support programs for SOs should incorporate this kind of parental intervention to improve circumstances for affected children.

Although most SOs described feeling concern about how the children might be affected, the consequences of the co-parent’s consumption were not perceived as severe enough to cause the children serious harm. This result can be compared to a previous Swedish qualitative study [40] where 23 (female) SOs sharing a child with a (male) co-parent with UAU were interviewed. In that study, all but two of the SOs had separated from their co-parents and stated that the reasons for this were severe substance use and major parental neglect, involvement of social services, and children showing delinquent behaviors. The 23 SOs all identified as parents to children who fared poorly as a consequence of co-parent drinking, a characteristic that the SOs in our study did not seem to share. Based on this, we reason that the consequences of co-parent UAU must be rather severe for SOs to initiate treatment-seeking based primarily on wanting support for affected children. In the present study, a majority of the SOs were still in a relationship with their co-parent, and their descriptions of perceived needs focused mainly on their own well-being and strategies for dealing with or relating to the co-parent. This is considered important knowledge for the development of support programs that can attract SOs at an earlier stage of co-parent UAU.

Lastly, several SOs mentioned that they felt personalized feedback was missing. This desire for feedback is understandable, but must be related to the potential gains of offering a program that is available at all times and does not require staffing, which for some SOs can be enough—at least as an important first step.

A strength of this study is that it reached a population that, to our knowledge, has not previously been represented in the literature. Thus, the results presented here are important for future studies aiming to attract this population. However, it is possible that the decision to include only participants who completed two or more of the modules in the SPARE program led to a risk of survival bias in our results.

One limitation is that the sample of SOs in our study was ethnically homogenous, with a high level of education, and thus does not reflect the Swedish population. However, we consider the current sample representative of the group of SOs who participated in the RCT and believe the results of the interviews are transferable to the larger study population.

An aspect regarding credibility involves the authors’ perspectives and experiences. The first author is a clinical psychologist specialized in cognitive behavior therapy and was involved in all parts of the RCT. The second author has a background in social work and was not involved in the quantitative evaluation. The close involvement of the first author comes with a risk of bias, potentially leading to more positive interpretations of data than by a “neutral” observer. This risk was highlighted already from the start of the study and we tried to counteract it by continuously discussing the analyses and results back and forth between the authors.