How AI Search Is Recommending Healthcare and Medical Devices: Monthly Trends

Executive Summary

Abbott FreeStyle remains the category leader in October 2026 with 61.2% valid recommendation coverage, ahead of second-place Ascensia CONTOUR at 39.7%, a gap of 21.5 points. The category was stable this month, with no brand exceeding normal month-to-month variation and no streak broken at the leadership level. Abbott FreeStyle's 2.1-point gain from September's 59.1% sits within normal movement, and its lead remains the widest in the series to date.

LifeScan OneTouch was the strongest upward mover over the comparable period, rising from 10.9% valid recommendation coverage in July 2026 to 13.5% in October 2026, up 2.6 points. On a prior-month basis the same brand moved up 3.3 points from September's 10.2%, and its rank-one rate rose to 1.7% from 0.2% at baseline. Ascensia CONTOUR was the sharpest decliner over the same span, easing from 40.1% in July to 39.7% in October, down 0.4 points, a move within normal variation.

Abbott FreeStyle's lead over Ascensia CONTOUR widened from 15.5 points in July to 21.5 points in October, and its lead over Roche Accu-Chek widened from 27.1 to 33.1 points over the same period. Both gaps widened 6.0 points across three months. The benchmark shows a category where the top position is not contested, the second tier is holding roughly steady, and the trailing brands occupy very small bases.

Each monthly run begins with prompt-surface observations across the benchmark's defined AI/search surface universe. July 2026 began with 657 observations (469 unique questions); August 2026 with 645 (486 unique questions); September 2026 with 672 (524 unique questions); October 2026 with 681 (531 unique questions). Of October's total, 681 mentioned a tracked brand or competitor; 673 were relevant and 8 were irrelevant. The public metrics use the 363 October observations that survive both qualification stages, compared with 432 in July 2026.

Key Findings

AI Response Inconsistency Alerts

Questions This Section Answers

• What kinds of factual conflicts did AI platforms produce about glucose monitoring brands this month?
• Where did AI responses disagree about TRUE METRIX recall status, recall scope, and discontinuation?
• Which pricing, accuracy, and coverage claims about Abbott FreeStyle conflicted across platforms?

The benchmark detected 29 critical and high-severity factual inconsistencies across six AI platforms in the current measurement month: ChatGPT, Copilot, Gemini, AI Mode, AI Overviews, and Perplexity. Three tracked brands were involved. These conflicts are reported at high confidence and describe direct contradictions within the same question, which means buyers asking a single question on different surfaces are receiving incompatible answers.

Trividia Health

Twelve high and critical inconsistencies centered on the FDA recall status and scope of TRUE METRIX blood glucose monitoring systems. When asked "Is TrueMetrix a good glucose meter?", AI Overviews stated that TRUE METRIX is widely considered an affordable, reliable, and accurate option with no mention of any recall or safety issue, citing Trividia Health's own product page, an Amazon listing, and a diabetes supply retailer. ChatGPT, answering the same question, stated that as of 2026 the FDA had issued a Class I recall and safety communication covering TRUE METRIX systems and recommends transitioning to an alternative testing method, citing the FDA safety communication and an FDA device registry entry.

Across the recall-scope conflicts, AI platforms provided conflicting information about whether the recall covers meters only or meters and test strips. When asked "Are True Metrix test strips being discontinued?", Copilot stated that the FDA announced all True Metrix meters and strips, including True Metrix Air, GO, and PRO, are affected by a recall due to a software design flaw, citing the FDA safety communication, the FDA recall alert, and MedTech Dive. Gemini, on the same question, stated that the FDA Class I recall and software correction is for TRUE METRIX meters due to an E-5 error display issue, not the test strips, citing a community health center notice. ChatGPT stated that the company indicated the meter, test strips, and control solution were not impacted. AI Mode stated that Trividia Health announced a labeling and software correction regarding an E-5 error code on True Metrix meters but characterized it as a safety and instruction update rather than a product discontinuation, citing Trividia Health's product notice page.

On discontinuation status, AI platforms diverged directly. When asked "Are true metrix test strips discontinued?", AI Mode stated that TRUE METRIX test strips are not discontinued, citing the FDA safety communication and Trividia Health notice pages. Perplexity stated that certain True Metrix test strips have been discontinued in some markets, citing the NDSS product schedule notice and Trividia Health's product page. The responses differed on availability in a way that affects purchase and switching decisions.

Abbott FreeStyle

Sixteen high-severity inconsistencies were detected across pricing, accuracy, availability, and coverage topics. AI platforms provided conflicting information about the monthly cost of FreeStyle Libre 2 without insurance. When asked "How much does the FreeStyle Libre 2 Sensor cost per month?", ChatGPT stated that cash prices average about $235 for two sensors, roughly $470 per month, citing Abbott's cost page and SingleCare. Copilot stated that retail price is about $235 for two sensors, which it described as $235 per month, and in separate responses stated uninsured monthly costs of $75 to $150 and annual costs of $1,400 to $3,300. The same $235 two-sensor figure was converted into materially different monthly and annual totals across responses.

On accuracy, responses differed on which device leads. When asked "What's the most accurate blood sugar monitor?", Gemini stated that Dexcom G7 is the top pick with the lowest MARD at 7.7% to 8.2% and that FreeStyle Libre 3 Plus is a strong competitor with comparable accuracy around 8.2%. Copilot stated that FreeStyle Libre 3 is the most accurate CGM at 7.8% to 7.9% MARD, better than Dexcom G7 at 8.2%, citing MDS Diabetes, GlucoseIntel, and BetterVitals. AI Mode stated that Dexcom G7 is the best overall for clinical accuracy, while Copilot in a different response stated that Abbott FreeStyle Libre 3 Plus is the most accurate at roughly 7.8% MARD.

On over-the-counter availability, responses conflicted. When asked "What is the most accurate CGM OTC?", ChatGPT stated that FreeStyle Libre 3 Plus is a prescription system, not OTC, and that Abbott's OTC option is Libre Rio, citing a device review article and clinical standards. Perplexity stated that Abbott FreeStyle Libre 3 is sold OTC in some areas as a consumer option, citing BetterVitals and a type 2 diabetes comparison site.

On real-time alerts, responses differed. When asked "What are the best continuous glucose monitoring devices?", Gemini stated that FreeStyle Libre 3 and Libre 3 Plus have alarms and alerts, citing a comparison guide. Copilot stated that FreeStyle Libre 3 Plus has no real-time alerts unless paired with an app, citing retailer product pages.

On Medicare coverage pathway, responses conflicted. When asked "What is the monthly cost of FreeStyle Libre 2?", AI Mode stated that Medicare coverage varies by specific Part D or Medicare Advantage copays, citing GoodRx and SingleCare. Copilot stated that the product is covered under Part B as durable medical equipment, citing a FreeStyle Libre cost site and Abbott support pages.

Ascensia CONTOUR

One high-severity inconsistency was detected on accuracy. When asked "Which blood sugar test kit is more accurate?", Gemini stated that Contour Next meters routinely test at a MARD of around 5.6%, citing a meter review site and diaTribe research on meter accuracy. Copilot stated that Contour Next ONE and Next GEN results fall within plus or minus 8.4% of lab values, citing retailer product listings. These are incompatible measurements of the same meter's accuracy.

Benchmark Context

Questions This Section Answers

• Why did the qualified observation count fall from 432 to 363 while the raw prompt collection grew?
• What is the difference between the collected prompt universe and the qualified benchmark set?

The report separates the raw collection universe from the qualified analysis set. Brand-level recommendation percentages are calculated within the qualified benchmark set.

The qualified observation count declined from 432 in July to 391 in September before settling at 363 in October, even as the raw collection universe expanded from 657 to 681 prompts. The gap between the two reflects a growing reserved pool rather than a shrinking category.

Benchmark-Level Metrics

The recommendation-shaped answer share peaked at 36.9% in August and the valid recommendation shortlist share rose every month from 56.5% in July to 59.6% in September and 63.6% in October. The category is producing recommendation-style answers at a higher rate even as the qualified set contracts.

AI Recommendation Trend

Questions This Section Answers

• Which brand leads the category, and how wide is the gap to second place?
• Which brand moved most over the comparable period, and where did it land in the standings?
• Did any brand's movement exceed normal month-to-month variation?

A stable leadership structure with a widening top gap and a small-base riser

The healthcare and medical devices category shows an uncontested leader and a clear second-place brand, with no change to the competitive structure in October. Gate status is stable: no brand exceeded normal month-to-month variation across the comparable period.

The category-level change over the comparable period reflects a combination of several smaller movements rather than a single dominant shift. No brand exceeded normal month-to-month variation from July to October. LifeScan OneTouch posted the largest rank-one gain in the category, rising from 0.2% to 1.7%.

What Changed This Month

Abbott FreeStyle

Abbott FreeStyle holds the category lead with valid recommendation coverage of 61.2% in October, up 2.1 points from September's 59.1% and up 5.6 points from July's 55.6%. The brand recorded 222 valid recommendations out of 363 qualified observations in October, compared with 231 out of 391 in September.

Top-three placement edged up from 51.7% in September to 53.7% in October, and rank-one rate rose from 9.7% to 14.6%. Raw mention presence eased from 95.7% in September to 91.5% in October, a 5.3-point decline from the July baseline of 96.8%.

The distinction for Abbott FreeStyle is that its recommended coverage and rank-one position are strengthening even as its raw mention presence softens. The brand is being recommended more often in the answers where it appears, but it is appearing in a slightly smaller share of them.

Highest-priority diagnostic: Which prompt types or surfaces account for the decline in raw mentions while recommendation credit rises?

LifeScan OneTouch

LifeScan OneTouch was the strongest riser with valid recommendation coverage of 13.5% in October, up 2.6 points from July's 10.9% and up 3.3 points from September's 10.2%. The brand recorded 49 valid recommendations out of 363 qualified observations in October, compared with 40 out of 391 in September.

Top-three placement rose from 7.2% in July to 10.7% in October, and rank-one rate moved from 0.2% to 1.7%, a 1.5-point gain. The brand appeared in 82 of 363 observations in October, up from 70 of 391 in September.

The distinction for LifeScan OneTouch is that it is gaining on every placement measure at once, from presence through to rank-one, even though its absolute counts remain modest.

Highest-priority diagnostic: Which prompts and surfaces produced the rank-one gain, and is the movement concentrated in a specific question type?

Ascensia CONTOUR

Ascensia CONTOUR holds second place with valid recommendation coverage of 39.7% in October, down 0.4 points from July's 40.1% and down 3.3 points from September's 43.0%. The brand recorded 144 valid recommendations out of 363 qualified observations in October, compared with 168 out of 391 in September.

Top-three placement eased from 39.1% in July to 37.5% in October, and rank-one rate moved from 28.0% to 25.6%. Raw mention presence held essentially flat at 51.5% in October versus 51.2% in July. The brand's average recommended rank of 1.6294 remains the best in the category.

The distinction for Ascensia CONTOUR is that its rank-one strength is the strongest in the field even as its overall coverage softens. The benchmark shows a brand that places first when it is recommended, but is recommended across a slightly smaller share of the qualified set.

Highest-priority diagnostic: Which prompts or surfaces account for the September-to-October coverage pullback while rank-one leadership holds?

Roche Accu-Chek

Roche Accu-Chek holds third place with valid recommendation coverage of 28.1% in October, down 0.4 points from July's 28.5% and down 2.3 points from September's 30.4%. The brand recorded 102 valid recommendations out of 363 qualified observations in October, compared with 119 out of 391 in September. This is the second consecutive month of decline.

Top-three placement edged up slightly from 20.8% in July to 21.2% in October, and rank-one rate held at 1.1% versus 1.2%. Raw mention presence was essentially flat at 39.4% in October versus 38.9% in July.

The distinction for Roche Accu-Chek is that its placement metrics are holding while its coverage drifts lower, a pattern that suggests the brand is maintaining position within the answers it appears in rather than losing ground within them.

Highest-priority diagnostic: Which prompt categories or surfaces account for the two-month coverage drift?

Prodigy Diabetes Care

Prodigy Diabetes Care has valid recommendation coverage of 1.4% in October, up 0.5 points from July's 0.9% but down 1.4 points from September's 2.8%. The brand recorded 5 valid recommendations out of 363 qualified observations in October, compared with 11 out of 391 in September. This is the second consecutive month of decline from August's series high of 3.0%.

Top-three placement rose from 0.2% in July to 1.1% in October. The brand's rank-one rate remained at zero across all four months. It appeared in 7 of 363 observations in October. These counts remain very small, so the movement should be read as directional rather than definitive.

Highest-priority diagnostic: Which prompts drove the August peak, and are the same prompts still active in October?

Trividia Health

Trividia Health has valid recommendation coverage of 0.8% in October, up 0.3 points from July's 0.5%, its series high. The brand recorded 3 valid recommendations out of 363 qualified observations in October, compared with 2 out of 391 in September. Top-three placement held at 0.5% and rank-one rate remained at zero across all four months. The brand appeared in 8 of 363 observations in October.

Abbott FreeStyle records 61.2% on valid recommendation coverage, while Trividia Health stands at 0.8%, a gap of 60.4 points. The more important issue is what that gap means for Trividia Health's recommendation position: Abbott FreeStyle's rank-one rate has climbed to 14.6% and Ascensia CONTOUR holds the category's strongest rank-one rate at 25.6%, while Trividia Health has recorded a 0% rank-one rate in every month of the series, meaning the brand has not been placed first in a single qualified recommendation to date. The small presence the brand does have — 8 of 363 October observations — is not converting into recommendation at any meaningful rate, and this is occurring in the same month the benchmark recorded twelve high and critical inconsistencies about TRUE METRIX's FDA recall status and discontinuation scope, meaning buyers researching the brand are encountering direct safety and availability contradictions on the same surfaces where Trividia Health is least visible in recommendation terms. Left unexamined, the combination of near-zero recommendation conversion and unresolved recall-messaging conflicts leaves the commercial cause of the brand's position unknown.

Highest-priority diagnostic: Which few prompts still surface the brand, and do the TRUE METRIX recall and discontinuation inconsistencies appear on those same prompts, suppressing conversion from presence to recommendation?

Buyer-Intent Interpretation

Questions This Section Answers

• Which buyer-intent cluster did October's qualified observations fall into, and which clusters registered none?
• What can the public benchmark not yet say about pricing, value, and brand-comparison questions?

All qualified observations in October fell into the Brand Recommendation cluster, with 363 observations assigned to this class. The Pricing & Value and Multi-Brand Comparison clusters registered no qualified observations this month, consistent with the pattern across the full series.

The public benchmark in October can answer which brands AI systems recommend most often and where they place. It cannot yet answer how AI systems frame pricing, value, or direct brand comparisons within the qualified set. The commercial picture rests on answer presence and position, not on cost signals or head-to-head framing. This gap is notable in a category where the inconsistency findings show pricing and comparison questions are being asked and answered, but outside the qualified benchmark denominator.

Brand Opportunity Summary

Questions This Section Answers

• Which diagnostic question is highest priority for each brand based on its October signal?
• Which brands show rising signals versus consecutive-month declines?

The benchmark identifies where attention is warranted across the category; a company-level analysis is needed to explain why those patterns hold. The inconsistency findings above indicate that context outside the qualified set, particularly on pricing, comparison, and safety topics, is also shaping what buyers see.

Evidence Behind the Benchmark

The aggregate metrics are built from prompt-level observations, including the query, the AI surface, the recommendation outcome, the rank position, sentiment, and citations where exposed. Company-level analysis can go deeper into prompt, competitor, surface, and evidence patterns to understand the drivers behind the headline coverage figures. Source presence is not automatically treated as proof of causation, and citation frequency is not endorsement.

About This Benchmark

This report is part of the LLM Authority Index AI Visibility Market Discovery research program.

• AI Visibility Industry Market Methodology
• AI Visibility Industry Market Metrics
• AI Visibility Industry Market Standards

Report-Specific Interpretation Notes

• Small-count movement: Prodigy Diabetes Care and Trividia Health operate on very small absolute counts, so their changes should be read as directional rather than definitive.
• Qualified denominator vs raw collection: All brand percentages are calculated against the 363 qualified observations, not the 681 prompts collected.
• Directional analysis: Month-over-month movement identifies changes worth investigating; it does not by itself establish the cause of those changes.

Next Step

The Public Benchmark Shows Where a Brand Is Winning or Losing. A Company-Level Audit Shows Why.

Beneath the aggregate percentages sit the questions that matter commercially: which high-intent prompts are won or lost, which competitor takes the recommendation when a brand loses, what attributes AI systems associate with each option, and which external sources shape those answers. The inconsistency findings in this report show that surface-level answers can diverge on recall status, pricing, and accuracy within the same question.

A company-specific AI visibility audit maps those prompt, surface, competitor, ranking, sentiment, and evidence-source patterns into a prioritized visibility strategy, turning the headline coverage figures into actionable direction.

Request an AI visibility audit