Incidence of disease after vasectomy: a record linkage retrospective cohort study. (Papers)

Introduction

Vasectomy is now a common method of permanent birth control. Though widely considered to have no adverse long term effects on health, doubts have occasionally been raised about its safety. These include concerns about the functional importance, if any, of the autoimmune sperm antibody response that commonly follows vasectomy[1 2] and the possibility of long term effects of the operation on the cardiovascular[3-10] and genitourinary systems. Also, there is current interest in the hypothesis that vasectomy might predispose men to cancers of the testis and prostate.[11-16] We report a medical record linkage study to identify the subsequent occurrence of these cancers and several other diseases after vasectomy.

Methods

We used data from the Oxford record linkage study, which includes brief abstracts of records of hospital inpatient care, day case care, and death certificates organised such that successive records relating to the same person can be linked together.[17 18] The data are coded from medical records by clerks trained by the staff of the study. Completeness of recording is routinely monitored.[19] Checks against independent sources of data have shown the data to be reliable.[20 21]

We analysed all records of men who underwent vasectomy between 1970 and 1986 and were aged 25-49 years and resident in the study area at the time of operation. Comparison cohorts comprised men in the same age group who had been admitted to hospital during the same period for one of three elective operations (removal of sebaceous cysts, removal of nasal polyps, or meniscectomy); for appendicitis; or for injuries (head injury or fractures of the humerus or femur). Each man was selected for one entry cohort only-that is, a man with a vasectomy and a comparison condition was included in the vasectomy cohort and excluded from the other cohort. Men with more than one eligible comparison condition were included only for the first condition in the rank order given in table 1. Records of hospital admissions and death after the index admission were sought to 31 December 1986 for the first occurrence of cancer of the testis, cancer of the prostate, leukaemia and other neoplasms of haematopoetic and lymphatic tissue, several autoimmune or possibly autoimmune diseases, and the cardiovascular conditions specified in table II. For each of these diagnoses men were excluded if the diagnosis had been recorded at the time of the index admission.

[TABULAR DATA OMITTED]

Age standardised first occurrence rates were calculated for each diagnosis …