No. More than 80 percent of the successful respondents took this step—but unsuccessful respondents were just as likely to take the same measure. That suggests that while cutting back on sugar may be a common approach to managing diabetes, it's not an essential one. Indeed, the American Diabetes Association now says that cutting back on sugar helps only if it's part of a comprehensive effort to manage total carbohydrate intake and especially weight. For more advice, see our tips on Eating right.

Yes. People who were successful in managing their diabetes were much more likely to cut back on calories than those who were unsuccessful (59 percent vs. 45 percent). That underscores the importance of weight loss in managing diabetes. Nearly three-quarters of the people in our survey said they were overweight before they learned they had diabetes. But 63 percent of the successful managers said they had lost weight since, compared with just 30 percent of those who were unsuccessful. For more advice, see our tips on Eating right.

Yes. Two-thirds of the people who were successful in managing their diabetes watched portion sizes, compared with barely half of those who were unsuccessful. That confirms our previous poll on weight loss, which found that carefully controlling portion sizes at each meal was strongly linked to having a lower body mass index, or BMI, a measure that correlates weight to height. People who lost weight were especially likely (62 percent) to report practicing portion control at least five days per week. So did 57 percent of the always thin, but only 42 percent of failed dieters. For more advice, see our tips on Eating right.

No. Roughly 40 percent of successful and unsuccessful respondents counted carbs, suggesting that this technique, while perhaps useful for some, isn't enough. It could also be that many people find this task—which involves carefully tallying the carbohydrate content of each component of their meal and then either limiting how much they eat or adjusting their medication dose—confusing and hard to stick with. To make it simpler, see our tips on Eating right.

No. Relatively few people in our survey tried the glycemic index, which ranks foods based on how quickly they raise blood glucose levels. Only 12 percent of people who successfully managed their diabetes used it, compared with 8 percent of those who weren't successful. While the glycemic index has been promoted as an effective dietary approach to diabetes, it's a complicated one to use in the real world. A food's effect on blood sugar levels depends on many factors, including how it's cooked and what it's consumed with. For example, eating even a higher-glycemic-index carbohydrate such as mashed potatoes with a little fat and protein will slow down glucose absorption. And many high-glycemic foods, such as beets, carrots, and watermelon, are loaded with vitamins and minerals, so avoiding them makes little nutritional sense. Instead, the most important step is to focus on how many carbohydrates you eat, not what type.

Yes. People who were successful in managing their diabetes were more likely to read food labels than were those who were unsuccessful (74 percent vs. 66 percent). That makes sense, since label reading is an essential step in choosing foods that are low in calories, which, in turn, is key to weight loss. Moreover, label reading can help people count carbs and choose foods that are high in fiber and whole grains, both of which also help people control their blood sugar. It can also help with the dietary management of common cardiovascular problems that often accompany diabetes. Reading food labels, for example, can help people with high blood pressure choose foods low in sodium and people with high LDL (bad) cholesterol levels choose foods low in saturated fat. For more advice, see our tips on Eating right.