Individuals living with a partner with an alcohol use disorder (AUD) can experience significant psychological distress and use health care more than those without a partner with an AUD. However, the prevailing treatment system's focus on the partner and personal barriers limit these individuals from getting help for themselves. Preliminary work on a self-directed, web-based coping skills training program, Stop Spinning My Wheels (SSMW), shows promise in broadening available treatments for this population. In this study, we conducted a robust evaluation of SSMW primary outcomes. The study aims to test whether women with a partner with an AUD assigned to SSMW experienced a greater reduction in negative affect (depression and anger) (1) than a usual web care (UWC) control and (2) with brief phone coach support (SSMW+coach) rather than without (SSMW only) and (3) whether baseline negative affect moderated treatment effects. Women (mean age 45.7, SD 10.8 years; Black: 17/456, 3.7%; White: 408/456, 89.5%) were randomized to SSMW only, SSMW+coach, or UWC. Depression (Beck Depression Inventory-II) and anger (State-Trait Anger Expression Inventory 2-State Anger) were assessed at baseline, 12-week posttest, and 6- and 12-month follow-ups. Participants in all conditions decreased in depression from baseline to posttest and from baseline to follow-up; SSMW-only and SSMW+coach participants decreased in anger, but UWC participants did not. Compared to UWC participants, SSMW-only participants experienced greater anger reduction (P=.03), and SSMW+coach participants experienced a greater reduction in depression (P<.001) from baseline to posttest. However, from baseline to follow-up, only a greater, but not statistically significant (P=.052), reduction in anger occurred in SSMW+coach compared to UWC. Although the SSMW conditions did not differ from each other in negative affect outcomes (P=.06-.57), SSMW+coach had higher program engagement and satisfaction (all P<.004). Baseline negative affect did not moderate effects, although remission from baseline clinically relevant depressive symptoms (Beck Depression Inventory≥14) was higher in SSMW only (33/67, 49%; odds ratio 2.13, 95% CI 1.05-4.30; P=.03) and SSMW+coach (46/74, 62%; odds ratio 3.60, 95% CI 1.79-7.23; P<.001) than in UWC (21/67, 31%); remission rates did not differ between the SSMW conditions (P=.12). The results partially supported the hypotheses. The SSMW conditions had earlier effects than UWC, but positive change in UWC mitigated the hypothesized long-term SSMW-UWC differences. The results highlight the importance of incorporating active controls in web-based clinical trials. Although SSMW+coach showed benefits over SSMW only on engagement and satisfaction measures and in the number needed to treat (5.6 for SSMW only; 3.2 for SSMW+coach), the SSMW conditions were comparable and superior to UWC on depressive symptom remission levels. Overall, SSMW with or without a coach can reduce clinically meaningful distress and add to available treatment options for this large, underserved group. ClinicalTrials.gov NCT02984241; https://www.clinicaltrials.gov/study/NCT02984241.