They choose to do it
Parents who choose to bed-share cite a variety of reasons. In one UK study Ball (Ball 2002) reported that 72% of babies who were breastfed for a month or more were at least occasional bed-sharers compared to 38% of babies who had never breastfed. Mothers in the UK and around the world have identified 'ease and convenience of breastfeeding' as an overwhelming reason for sleeping with their infants at night (Ball 2002; Baddock et al 2007; McKenna & Volpe 2007; Ateah & Hamelin 2008). For more information on breastfeeding and bed-sharing see here.
Other reasons included the enjoyment of close contact with their baby; anxiety regarding their baby's health; ease of settling an irritable baby; and a family bed parenting philosophy. Studies in the US and Canada found that mothers reported bed-sharing despite being told not to, because the risks seemed "to be outweighed by the perceived benefits and convenience" (Ateah & Hamelin 2008: 279) and "their sleep arrangement was the only way that worked for them" (Kendall-Tacket et al 2010: 30).
They have no alternatives
In the study by Ball (2002) some parents living in impoverished circumstances slept with their newborn out of necessity rather than choice, being unable to afford a cot, or the space to house a cot. This has been shown to be a particular issue amongst African American families of low socio-economic status. Joyner et al’s (2010) study found that in addition to space issues, bed-sharing was regarded as protective against environmental dangers arising from poor housing conditions (e.g. vermin) or living in high crime areas. The association between poverty and bed-sharing in some groups, drives programmes which aim to provide all infants with a cot to sleep in.
It happens accidentally
Many of the tragic outcomes associated with bed-sharing seem to have occurred when parents fall asleep with their babies without intending to, and without ensuring that the environment is as safe as possible for the baby. In the vast majority of such cases a parent or care-giver has been under the influence of drugs or alcohol while in charge of a baby. A recent study by Blabey and Gessner (2009) examined 13 years of data on Alaskan infant deaths while bed-sharing. In 99% of cases at least one additional risk factor was present. The most common of these was maternal tobacco use (75%), but this was followed by sleeping with an impaired person (43%). Blair et al (2009)? found that 25% of babies who died while co-sleeping were doing so with an adult who had consumed 2 or more units of alcohol.
Recommendations advising parents to ensure their babies always sleep alone in a cot are aimed at eliminating these sorts of tragedies by attempting to ensure that babies can never fall asleep with a care-giver who is inattentive or even unconscious. The increase in deaths while sofa-sharing (Blair et al 2009) may be partly due to parents leaving the bed-room to feed or settle a baby, and falling asleep accidentally on a sofa or couch.