Exit Interviews Reveal What Patients Really Felt About Tirzepatide and Dulaglutide

For patients with type 2 diabetes and established heart disease, the numbers from clinical trials tell only part of the story. HbA1c percentages, kilograms lost, and cardiovascular event rates are the currency of regulatory science, but they say little about how a treatment actually feels from the inside. A new qualitative study published in Diabetes Therapy has now filled that gap for two of the most widely discussed glucose-lowering therapies of the modern era, interviewing 74 people who spent nearly four years taking either tirzepatide or dulaglutide in the landmark SURPASS-CVOT trial. Their accounts reveal a treatment experience that extended far beyond laboratory values, touching energy levels, daily routines, emotional well-being, and even long-standing joint pain.

The research team, led by Katie D. Stewart and Louis S. Matza of PPD Evidera Patient-Centered Research at Thermo Fisher Scientific, together with colleagues at Eli Lilly and Company, conducted semi-structured exit interviews after participants had fully completed the multiyear SURPASS-CVOT trial. That phase 3, randomized, double-blind, noninferiority study compared tirzepatide, a dual agonist of the GLP-1 and GIP receptors, with dulaglutide, a pure GLP-1 receptor agonist, in patients with type 2 diabetes and atherosclerotic cardiovascular disease. Because treatment assignment remained blinded until the interviews and analyses were finished, participants spoke about their experience without knowing which drug they had received, and the researchers deliberately presented results for the combined sample.

Recruitment took place across ten US trial sites in California, Florida, Illinois, Kansas, Maryland, and Texas. Of 147 people invited, 74 completed interviews between August and October 2025, each lasting roughly 90 minutes by telephone or web conference. The participants were predominantly male at 63.5 percent, with a mean age of 68.9 years, and had lived with type 2 diabetes for an average of 18.6 years. Most carried substantial cardiovascular histories: nearly 80 percent reported a prior heart attack or heart disease, and hypertension and arthritis were each present in roughly 45 percent of the sample. On average, they had received the blinded study medication for 45.0 months, and during that time their weight fell by a mean of 14.6 kilograms while HbA1c dropped 1.7 percentage points, from 8.3 percent to 6.6 percent.

The interview methodology was rigorous by qualitative standards. Six trained female interviewers used a guide that began with open-ended questions about changes participants noticed, then probed systematically for specific concepts such as weight, blood sugar, energy, appetite, sleep, and joint pain. Transcripts were coded in ATLAS.ti by seven trained coders who achieved greater than 90 percent agreement before coding independently, and the team formally documented saturation: fewer than 5 percent of concepts first appeared in the final 21 interviews, and no new concepts emerged in the last 11. That statistical-style discipline matters, because it signals that the 74 voices captured here likely represent the full range of experiences among long-term users of these drugs.

The headline finding is the sheer breadth and consistency of perceived benefit. Ninety-nine percent of participants reported at least one treatment benefit, and 97.3 percent described improved glycemic control, most often citing lower HbA1c but also steadier day-to-day blood glucose. Weight reduction was reported by 91.9 percent, with 20 participants characterizing the loss as substantial using words like dramatic or drastic, and most noticing the change within the first six months. Reduced appetite was described by 78.4 percent, frequently as a curbed or suppressed drive to eat and a new sense of fullness after smaller portions. Nearly two-thirds reported improved diet and eating habits, and 41.9 percent said their desire for unhealthy food had diminished. More than half reported increased energy, and a striking number volunteered improvements in conditions beyond diabetes, including joint pain or osteoarthritis in 18.9 percent and sleep apnea in 8.1 percent.

Those physical changes translated into lived consequences. Eighty-five percent of participants said the treatment affected their quality of life, with 81 percent describing that impact as positive, and 77 percent reported changes in at least one domain of daily activity, most commonly physical activities, followed by activities of daily living, family activities, social life, leisure, and work. Seventy percent said the treatment changed their interactions with others, often through comments on their appearance or weight loss. Thirty-nine percent reported emotional effects, most frequently feeling happier, more confident, more motivated, or less worried, and these were usually tied to the weight loss or steadier glucose. Even health behaviors shifted: among participants who drank alcohol, ten reported drinking less or wanting it less, and four of thirteen nicotine users reported smoking less or quitting altogether.

The safety picture was familiar but not trivial. Fifty-four participants reported at least one adverse event, with gastrointestinal symptoms dominating: nausea was the most common at 28.4 percent, followed by diarrhea, stomach upset, vomiting, constipation, and reflux. Six participants experienced low blood sugar episodes, and four described appetite suppression so severe that food became repulsive, with one tirzepatide-treated participant recalling that seeing food made him sick and he knew he was really sick because he always loved to eat. Two participants lost more weight than they wanted, and one 86-year-old man reduced his own dose after deciding the weight loss had gone too far. Other reported events included fatigue, insomnia, body soreness, muscle loss, injection-site reactions, and chills. Yet despite these burdens, 93 percent said they would be very likely or likely to continue the medication if it were available, and 93 percent would recommend it to others.

An exploratory subgroup analysis hinted at differences between the two drugs, though the study was never designed to compare them statistically. Benefits trended somewhat more frequently among the 43 tirzepatide-treated participants than the 31 on dulaglutide: improved glycemic control was reported by 100 percent versus 94 percent, weight reduction by 95 percent versus 87 percent, reduced appetite by 86 percent versus 68 percent, improved eating habits by 74 percent versus 52 percent, and increased energy by 67 percent versus 42 percent. Similar patterns appeared for quality-of-life impacts, including comments from other people, which 67 percent of tirzepatide patients reported versus 32 percent of dulaglutide patients. Whether these gaps reflect true pharmacological differences, the dual GIP activity of tirzepatide, or simply sample composition remains an open question for larger dedicated comparisons.

What distinguishes this study from its predecessors is duration. Previous qualitative investigations of GLP-1-based therapies, including exit interviews from tirzepatide trials in diabetes, weight management, and obstructive sleep apnea, and a study of the triple agonist retatrutide, covered treatment periods of one year or less. Here, patients averaged 45 months on therapy, and their enthusiasm did not fade with time. The authors argue this is the first qualitative study to assess patient perceptions of GLP-1 receptor agonist-based treatment over multiple years, and the persistence of reported benefits suggests that the improvements patients describe in shorter trials are not a honeymoon effect but a durable feature of long-term treatment.

The authors are careful about the limits of their data. All participants had type 2 diabetes with elevated cardiovascular risk, so generalizability to broader diabetes populations is uncertain, and only US sites were interviewed despite the multinational trial. Selection bias may favor healthier or more satisfied patients, and only five of the 74 had discontinued treatment early, so the perspectives of people who quit these drugs are largely missing. Recall bias is also a concern, since interviews occurred an average of 6.1 months after the last dose because they were delayed until the trial database was locked. Most importantly, the researchers caution that improvements in sleep, osteoarthritis, cardiovascular health, alcohol use, and nicotine use may be secondary consequences of weight loss rather than direct drug effects, and the small numbers reporting them warrant caution. Still, the core message stands: when patients who lived with these medications for years are asked what mattered, they describe transformations in blood sugar, body weight, appetite, energy, and daily life that the trial’s quantitative endpoints could only partially capture, and nearly all of them would choose the treatment again.

Subject of Research: Patient-reported experiences of long-term treatment with tirzepatide and dulaglutide in adults with type 2 diabetes and atherosclerotic cardiovascular disease, assessed through qualitative exit interviews after the SURPASS-CVOT trial.

Article Title: Exit Interviews Examining Patient Experiences with Tirzepatide and Dulaglutide for Treatment of Type 2 Diabetes in the SURPASS-CVOT Trial

Article References: Stewart, K. D., Matza, L. S., Chinthammit, C., Pavo, I., Bartee, A. K., & Boye, K. S. (2026). Exit Interviews Examining Patient Experiences with Tirzepatide and Dulaglutide for Treatment of Type 2 Diabetes in the SURPASS-CVOT Trial. Diabetes Therapy. https://doi.org/10.1007/s13300-026-01907-y

Image Credits: AI Generated

DOI: 10.1007/s13300-026-01907-y

Keywords: tirzepatide, dulaglutide, type 2 diabetes, GLP-1 receptor agonist, SURPASS-CVOT, exit interviews, qualitative research, patient-reported outcomes, weight loss, glycemic control, quality of life, cardiovascular disease