Does Cosleeping Contribute to Lower Testosterone Levels in Fathers? Evidence from the Philippines

Abstract

Because cross-species evidence suggests that high testosterone (T) may interfere with paternal investment, the relationships between men's transition to parenting and changes in their T are of growing interest. Studies of human males suggest that fathers who provide childcare often have lower T than uninvolved fathers, but no studies to date have evaluated how nighttime sleep proximity between fathers and their offspring may affect T. Using data collected in 2005 and 2009 from a sample of men (n = 362; age 26.0 ± 0.3 years in 2009) residing in metropolitan Cebu, Philippines, we evaluated fathers' T based on whether they slept on the same surface as their children (same surface cosleepers), slept on a different surface but in the same room (roomsharers), or slept separately from their children (solitary sleepers). A large majority (92%) of fathers in this sample reported practicing same surface cosleeping. Compared to fathers who slept solitarily, same surface cosleeping fathers had significantly lower evening (PM) T and also showed a greater diurnal decline in T from waking to evening (both p<0.05). Among men who were not fathers at baseline (2005), fathers who were cosleepers at follow-up (2009) experienced a significantly greater longitudinal decline in PM T over the 4.5-year study period (p<0.01) compared to solitary sleeping fathers. Among these same men, baseline T did not predict fathers' sleeping arrangements at follow-up (p>0.2). These results are consistent with previous findings indicating that daytime father-child interaction contributes to lower T among fathers. Our findings specifically suggest that close sleep proximity between fathers and their offspring results in greater longitudinal decreases in T as men transition to fatherhood and lower PM T overall compared to solitary sleeping fathers.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 68 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.