Fig. 3: Logistic regression analyses reveal expected and less expected diagnoses associated with male infertility that were first obtained before the 6-month cutoff. | Communications Medicine

Fig. 3: Logistic regression analyses reveal expected and less expected diagnoses associated with male infertility that were first obtained before the 6-month cutoff.

From: Leveraging electronic health records from two hospital systems identifies male infertility-associated comorbidities across time

Fig. 3

a Volcano plot of patients’ diagnosis associations with male infertility for diagnoses first obtained before the 6-month cutoff at UC. b Volcano plot of patients’ diagnosis associations with male infertility for diagnoses first obtained before the 6-month cutoff at Stanford. c Truncated Manhattan plot of diagnoses that were first obtained before the 6-month cutoff at UC, grouped into phecode categories. The following four diagnoses are not included in the plot: Abnormal spermatozoa (−log10(p-value) = 106.74); Persons encountering health services in circumstances related to reproduction (−log10(p-value) = 231.35); Testicular hypofunction (−log10(p-value) = 231.35); and Varicose veins (−log10(p-value) = 115.62). d Manhattan plot of diagnoses that were first obtained before the 6-month cutoff at Stanford, grouped into phecode categories. For panels (a) and (b), pink dots = −log10(p-value) of diagnoses significantly associated with male infertility; blue dots = -log10(p-value) of diagnoses significantly associated with vasectomy-related record; gray dots = -log10(p-value) of diagnoses that are not significantly associated with male infertility status. Phenotype corresponds to diagnosis. For (c), dots indicate the −log10(p-value) of diagnoses organized by phecode category, where dark red dots—circulatory system, black dots—congenital anomalies, blue-green dots—dermatologic, gold dots—digestive, brown dots—endocrine/metabolic, blue dots—genitourinary, orange dots—hematopoetic, red-orange dots—infectious disease, dark purple dots—injuries & poisonings, dark green dots—mental disorders, tan dots—musculoskeletal, light purple dots—neoplasms, light green dots—neurological, red dots—null, dark brown dots—pregnancy complications, reddish-brown dots—respiratory, pink dots—sense organs, and medium green dots—symptoms. For panel (d), dots indicate the −log10(p-value) of diagnoses organized by phecode category, where dark red dots—circulatory system, black dots—congenital anomalies, blue-green dots—dermatologic, gold dots—digestive, brown dots—endocrine/metabolic, blue dots—genitourinary, orange dots—hematopoetic, red-orange dots—infectious disease, dark purple dots—injuries & poisonings, dark green dots—mental disorders, tan dots—musculoskeletal, light purple dots—neoplasms, light green dots—neurological, red dots—pregnancy complications, dark brown dots—respiratory, reddish-brown dots—sense organs, and pink dots—symptoms. For panels (a, c), 1442 diagnoses were compared at UC, and Benjamini-Hochberg correction was used to identify diagnoses significantly associated with male infertility. For panels (b, d), 1197 diagnoses were compared at Stanford, and Benjamini-Hochberg correction was used to identify diagnoses significantly associated with male infertility. For all panels, top diagnoses found to be significantly associated with male infertility in the primary analysis are annotated. Phecode categories correspond to disease categories. n = 6531 UC patients with male infertility; n = 8353 UC patients with vasectomy-related record; n = 5551 Stanford patients with male infertility, n = 2464 Stanford patients with vasectomy-related record. P—p-value, UC—University of California.

Back to article page