David developed type 1 diabetes after landing on his head in a bike accident (he was not wearing a helmet). He had no diabetes symptoms before the accident, but only days afterwards he returned to the ER and was diagnosed with type 1. It is likely that there are different causes of diabetes in different people; perhaps some type of physical stress triggered his diabetes? 

Type 1 diabetes

Sepa and Ludvigsson (2006) reviewed the literature concerning psychological stress and type 1 diabetes. They found that 9 of 10 studies found associations between stress and type 1 diabetes. Additionally, one large study found an association between stress and type 1-related autoimmunity at early ages in life in the general population. They conclude that psychological stress can accelerate the appearance of type 1 diabetes, and may also contribute to the induction or progression of type 1 diabetes-associated autoimmunity, but more research is needed. The mechanisms for these effects are not known, but may involve beta cell stress or direct influence on the immune system; psychological stress can also increase insulin resistance. Psychological stress in children is linked to changes in the immune system, as well as effects on beta cells (Carlsson et al. 2014).
An interesting study analyzed a large number of people in Denmark, and found that if mothers experienced severe bereavement during pregnancy, their offspring were more likely to develop type 1 diabetes. The link was strongest for traumatic death of a sibling or the father, not for non-traumatic death or deaths of grandparents. And, the effect was strongest in girls (Virk et al. 2010).
Major life events have also been associated with the onset of type 1 diabetes, possibly due to increased levels of stress hormones, which are also increased in conditions involving inflammation (such as type 1 diabetes) (Dahlquist 2006).

A study from Israel found that that trauma of war was associated with an increased risk of type 1 diabetes. Children living in regions that were attacked during the Second Lebanon War had a higher risk of type 1 in the four years after the war, as compared to those living in areas that were not attacked. The associations were strongest in boys (Zung et al. 2012).

The "Beta Cell Stress Hypothesis" suggests that any phenomenon that induces insulin resistance, and thereby extra pressure on the beta cells, should be regarded as a risk factor for type 1 diabetes (Ludvigsson 2006Sepa and Ludvigsson 2006). In the "Overload Hypothesis," Dahlquist (2006) also points out that various environmental factors, including physical or psychological stress, could overload beta cells, thus hastening the appearance of type 1 diabetes (see the hypotheses page for more on these hypotheses).

Type 2 diabetes

Stress has also been linked to type 2 diabetes. In a large German study, people who suffered from post-traumatic stress disorder were more likely to have type 2 diabetes than those who did not (Lukaschek et al. 2013). 

Laboratory studies

Chronic exposure to stress makes mice more susceptible to autoimmunity (Harpaz et al. 2013).

Mice exposed to chronic stress in combination with a high fat diet showed impaired glucose tolerance (Castaneda et al. 2011).

Stress and environmental contaminants

Environmental contaminants can exacerbate the effects of stress. For example, stress and air pollution can act synergistically, exacerbating respiratory disease (Clougherty and Kubzansky 2009). A combination of stress and contaminants may also affect offspring when parents are exposed. In one study, exposing rat mothers to either stress or the heavy metal lead altered the response to stress in their offspring, and the effects were greater when combined (Virgolini et al. 2006). Thus exposures to contaminants may be important when studying the effects of stress on type 1 diabetes development.

The bottom line

Stress may be able to accelerate the appearance of type 1 diabetes, but more research is needed.