ObjectiveTo quantify the risk of ectopic pregnancy among a transient diagnosis of pregnancy of unknown location (PUL).MethodsRetrospective cohort study between August 2016 and November 2020. The final cohort included 244 patients with 255 PULs who presented with vaginal bleeding and/or abdominal pain, positive serum quantitative β‐human chorionic gonadotropin, and negative transvaginal ultrasound, with a subsequent definitive diagnosis. Two‐way analysis of variance was used with significance set at P < 0.050. Bonferroni‐corrected significance values were used in post hoc analysis. Multinominal logistic regression was used to predict adjusted risk for pregnancy outcome.ResultsDefinitive diagnosis consisted of 101 (39%) intrauterine pregnancies, 33 (13%) ectopic pregnancies, and 121 (48%) resolved PULs or resolved/treated persisting PULs. A total 68% of the PULs subsequently became nonviable. Vaginal bleeding was associated with increased risk of nonuterine pregnancies.ConclusionA total of 13% of PULs were subsequently diagnosed as ectopic pregnancies, which is higher than the 2% to 3% risk of an ectopic pregnancy in the general reproductive population. A total of 68% of PULs were subsequently nonviable, which is higher than the 31% early pregnancy loss rate in the general reproductive population. This study quantitatively confirms that a transient diagnosis of a PUL increases the odds for ectopic pregnancy and early pregnancy loss.