How Well Do Artificial Intelligence Chatbots Respond to the Top Search Queries About Urological Malignancies?

Artificial intelligence (AI) chatbots are becoming a popular source of information but there are limited data on the quality of information on urological malignancies that they provide. Our objective was to characterize the quality of information and detect misinformation about prostate, bladder, kidney, and testicular cancers from four AI chatbots: ChatGPT, Perplexity, Chat Sonic, and Microsoft Bing AI. We used the top five search queries related to prostate, bladder, kidney, and testicular cancers according to Google Trends from January 2021 to January 2023 and input them into the AI chatbots. Responses were evaluated for quality, understandability, actionability, misinformation, and readability using published instruments. AI chatbot responses had moderate to high information quality (median DISCERN score 4 out of 5, range 2-5) and lacked misinformation. Understandability was moderate (median Patient Education Material Assessment Tool for Printable Materials [PEMAT-P] understandability 66.7%, range 44.4-90.9%) and actionability was moderate to poor (median PEMAT-P actionability 40%, range 0-40%The responses were written at a fairly difficult reading level. AI chatbots produce information that is generally accurate and of moderate to high quality in response to the top urological malignancy-related search queries, but the responses lack clear, actionable instructions and exceed the reading level recommended for consumer health information. PATIENT SUMMARY: Artificial intelligence chatbots produce information that is generally accurate and of moderately high quality in response to popular Google searches about urological cancers. However, their responses are fairly difficult to read, are moderately hard to understand, and lack clear instructions for users to act on.

Assessment of Artificial Intelligence Chatbot Responses to Top Searched Queries About Cancer

Importance

Consumers are increasingly using artificial intelligence (AI) chatbots as a source of information. However, the quality of the cancer information generated by these chatbots has not yet been evaluated using validated instruments.

Objective

To characterize the quality of information and presence of misinformation about skin, lung, breast, colorectal, and prostate cancers generated by 4 AI chatbots.

Design, Setting, and Participants

This cross-sectional study assessed AI chatbots' text responses to the 5 most commonly searched queries related to the 5 most common cancers using validated instruments. Search data were extracted from the publicly available Google Trends platform and identical prompts were used to generate responses from 4 AI chatbots: ChatGPT version 3.5 (OpenAI), Perplexity (Perplexity.AI), Chatsonic (Writesonic), and Bing AI (Microsoft).

Exposures

Google Trends' top 5 search queries related to skin, lung, breast, colorectal, and prostate cancer from January 1, 2021, to January 1, 2023, were input into 4 AI chatbots.

Main Outcomes and Measures

The primary outcomes were the quality of consumer health information based on the validated DISCERN instrument (scores from 1 [low] to 5 [high] for quality of information) and the understandability and actionability of this information based on the understandability and actionability domains of the Patient Education Materials Assessment Tool (PEMAT) (scores of 0%-100%, with higher scores indicating a higher level of understandability and actionability). Secondary outcomes included misinformation scored using a 5-item Likert scale (scores from 1 [no misinformation] to 5 [high misinformation]) and readability assessed using the Flesch-Kincaid Grade Level readability score.

Results

The analysis included 100 responses from 4 chatbots about the 5 most common search queries for skin, lung, breast, colorectal, and prostate cancer. The quality of text responses generated by the 4 AI chatbots was good (median [range] DISCERN score, 5 [2-5]) and no misinformation was identified. Understandability was moderate (median [range] PEMAT Understandability score, 66.7% [33.3%-90.1%]), and actionability was poor (median [range] PEMAT Actionability score, 20.0% [0%-40.0%]). The responses were written at the college level based on the Flesch-Kincaid Grade Level score.

Conclusions and Relevance

Findings of this cross-sectional study suggest that AI chatbots generally produce accurate information for the top cancer-related search queries, but the responses are not readily actionable and are written at a college reading level. These limitations suggest that AI chatbots should be used supplementarily and not as a primary source for medical information.

Association between pregnancy and nocturia: A National Health and Nutrition Examination Survey analysis

AIMS

This study aimed to determine the impact of pregnancy and pregnancy-associated characteristics on nocturia.

METHODS

Using the National Health and Nutrition Examination Survey (NHANES) database from 2005/06 to 2017/18, we analyzed women who completed the "Reproductive Health" and "Kidney Conditions-Urology" questionnaires by sorting them into three groups: nulligravida, previously pregnant (nonpregnant with prior pregnancies), and currently pregnant. After excluding patients with pre-existing conditions impacting nocturia or with incomplete data, we weighed and matched the groups for age, race, BMI, and number of pregnancies. The relationships of nocturia to pregnancy in all groups and pregnancy-associated characteristics (gestational diabetes mellitus [GDM], history of multiple pregnancies, and trimesters of pregnancy) in currently pregnant women were assessed.

RESULTS

Of 8330 women that indicated pregnancy status, 1544 women (age range: 20-44 years; 523 nulligravida, 498 previously pregnant, 523 currently pregnant women) were included in analysis. Currently, pregnant women had a higher prevalence of nocturia than previously pregnant and nulligravida women (56.4% vs. 22.5% vs. 16.1%, p < 0.001) and had the highest odds of nocturia (OR: 6.82, p < 0.001). GDM or history of multiple pregnancies showed no associations in currently pregnant women. Increasing trimesters were associated with nocturia, with the third trimester showing the highest odds (OR​​​: 10.35, p < 0.001) and a greater average of nighttime voids than the first and second trimesters (2.40 ± 1.42 vs. 1.56 ± 1.31 and 1.88 ± 1.32, p < 0.001).

CONCLUSIONS

The association noted between pregnancy and nocturia, which strengthened with increasing trimesters, demonstrates that nocturia can significantly impact quality of life and therefore must be addressed during pregnancy.

Impact of COVID-19 on Online Interest in Urologic Conditions: An Analysis of Google Trends

Background

With COVID-19 leading to several isolation measures for preventative care, health care utilization, especially within urology, decreased substantially. The impact of COVID-19 on the population’s interests in urologic conditions remains to be established. By using the platform of Google Trends, which allows search behaviors and interest in healthcare topics to be quantified over time, we investigated the impact of COVID-19 on online search behaviors relating to common urologic conditions in the US.

Methods

The platform of Google Trends was utilized to analyze online interest in twelve common urologic conditions in the US from October 1, 2018 to August 1, 2021 (divided into “pre-COVID” and “COVID” periods at March 1, 2020). Search volume index (SVI), a measure of relative search volume on Google, data sets for the US, top queried and populated states, rising queries, and top queries were retrieved and analyzed for all conditions. Pre-COVID and COVID median SVIs were compared using the Mann Whitney U test, and correlations were analyzed using Spearman’s rank-order correlation test.

Results

For all twelve urologic conditions, rising and top queries were often related to symptoms, treatments, and COVID-19. COVID showed higher SVIs for erectile dysfunction (p=0.04) and lower SVIs for bladder cancer (p<0.01), hematuria (p<0.01), kidney cancer (p<0.01), kidney stones (p=0.03), and prostate cancer (p<0.01). Correlations to COVID-19 searches were seen for bladder cancer (R_S=-0.36, p<0.01), erectile dysfunction (R_S=0.20, p=0.04), hematuria (R_S=-0.31, p<0.01), overactive bladder (R_S=-0.23, p=0.04), and prostate cancer (R_S=-0.33, p<0.01). No correlations were found for benign prostatic hyperplasia, interstitial cystitis, low testosterone, urinary incontinence, and urinary tract infections.

Conclusions

Online interest in many urologic conditions, especially cancers, decreased during COVID. Given the internet’s increasing role in healthcare, a reduced interest could translate to delayed diagnosis and treatment of these conditions. Only erectile dysfunction showed increasing interest, potentially due to research or misinformation linking it to COVID-19.

Kidney Stone Recurrence among Children and Adolescents

PURPOSE

Kidney stone disease has become increasingly common during childhood and adolescence. However, the rate of symptomatic kidney stone recurrence for pediatric patients is uncertain. We sought to determine the recurrence rate of symptomatic kidney stones in a cohort of children with incident symptomatic nephrolithiasis.

MATERIALS AND METHODS

We performed a retrospective cohort study of patients 3 to 18 years old without anatomical abnormalities or genetic causes of nephrolithiasis who presented with a first symptomatic kidney stone between 2008 and 2014. We determined recurrence rates of symptomatic nephrolithiasis, defined as a new kidney stone on ultrasound and/or computerized tomogram associated with pain and/or vomiting. We also estimated associations between completing 24-hour urinalysis and symptomatic kidney stone recurrence using Kaplan-Meier curves and multivariable Cox regression models.

RESULTS

A total of 285 children with a median age of 14.8 years (IQR 11.3-16.6) at nephrolithiasis diagnosis were followed for 492 person-years. A total of 86 symptomatic recurrent stones developed in 68 patients (24%) during the followup period. The probability of symptomatic stone recurrence was 50% at 3 years after the index kidney stone. Median time to stone recurrence was 3 years at the first recurrence and 5 years at the second. Adjusting for confounders including adherence to followup, completing a 24-hour urinalysis after a kidney stone episode was associated with a 60% decreased risk of recurrence (hazard ratio 0.40, 95% CI 0.18-0.91).

CONCLUSIONS

The risk of kidney stone recurrence is high during childhood, with approximately 50% of children presenting with symptomatic recurrence within 3 years of the first stone. The role and usefulness of analyzing 24-hour urine chemistries in decreasing kidney stone recurrence should be explored in future prospective studies.

Surgical Considerations for the Geriatric Urology Patient

INTRODUCTION

As the number of geriatric patients continues to increase, urologists will encounter more elderly patients who require counseling about medical or surgical management of their conditions. In this review we provide a practical pathway for the elderly patient being considered for urological surgery.

METHODS

Our review includes preoperative evaluation and assessment for cognition, frailty, functional status, falls, cardiovascular and pulmonary status, and nutritional state.

RESULTS

Intraoperative concerns include operative procedure choice (with minimally invasive approaches emphasized), positioning, hypothermia, and antibiotic and venous thromboembolism prophylaxis. Postoperative attention requires assessment for delirium, early ambulation, fall prevention, lung expansion, avoidance of nasogastric tubes, early oral feeding and adequate but age adjusted pain control. Appropriate discharge from the hospital must follow, with planning started in the preoperative phase.

CONCLUSIONS

Special attention to these adjustments in the operative pathway lead to high operative success rates with a lower risk of complications.

The presence of high-grade prostatic intraepithelial neoplasia or atypia on prostate biopsy does not adversely affect prostatectomy outcomes for patients otherwise eligible for active surveillance

OBJECTIVE

To investigate if the presence of concomitant high-grade prostatic intraepithelial neoplasia (HGPIN) or atypical small acinar proliferation (ASAP) on biopsy increases the risk of occult adverse pathology in patients otherwise suitable for active surveillance (AS).

METHODS

Patients with D'Amico low-risk prostate cancer on ≥ 10-core biopsy who underwent radical prostatectomy at our academic center were evaluated for eligibility for AS by either Epstein criteria or Memorial Sloan Kettering Cancer Center (MSKCC) criteria. Prostatectomy specimens of patients eligible for AS were compared to determine if the presence of clinical HGPIN or ASAP affected the primary outcomes of pathologic upstaging and Gleason score upgrading.

RESULTS

Of 553 patients with low-risk prostate cancer, 400 patients (72.3%) met the MSKCC criteria, whereas only 170 patients (30.7%) met the Epstein criteria. HGPIN was present in approximately 32%, and ASAP in approximately 12%, of each AS cohort. On univariate and multivariate analyses, HGPIN and ASAP had no impact on the rate of upgrading and upstaging in either Epstein or MSKCC AS-eligible patients. Furthermore, the presence of HGPIN and ASAP had no impact on the 5-year biochemical recurrence-free survival.

CONCLUSION

The presence of HGPIN or ASAP does not increase the risk of upgrading, upstaging, or adverse pathology at the time of prostatectomy for patients who meet the AS criteria. If otherwise suitable, HGPIN and ASAP should not impact the decision to choose AS. However, analysis of prospective AS trials is required to determine if HGPIN or ASAP impacts tumor progression once on AS.

Prostate Stromal Tumor of Uncertain Malignant Potential: Case Report With 5-Year Follow-up

Prostate stromal tumor of uncertain malignant potential is a term used to describe a specialized proliferation of stromal cells within the prostate. Most of these tumors tend to be benign, but some can present with local invasion or progress to prostatic stromal sarcoma with distant metastasis. We report a case of a 62-year-old male patient who presented to us with a diagnosis of stromal tumor of uncertain malignant potential. We have followed up the patient for 5 years with imaging, prostate-specific antigen checks, and annual prostate biopsies.

Simple, age-based formula for predicting renal length in children

OBJECTIVE

To determine a simple, age based formula for predicting ideal renal length in children. Renal size is a valuable marker in the evaluation of children with urological disorders. Although many authors have described complex nomograms and multivariate formulas for determining renal size, we propose a simple and accurate formula.

MATERIAL AND METHODS

All renal ultrasound (US) studies performed over a 9-year period in patients <18 years of age were retrospectively evaluated, excluding patients with a history of urinary tract disease or with abnormal renal US findings.

RESULTS

Ultrasounds were performed in 778 children <18 years who met inclusion criteria. Sixty-one percent of the patient population was ≥1 year of age at the time of the US. Forty-four percent of the children were male. In children 1 year of age or older, the formula was length (cm) = age (years) × 0.3 + 6, R(2) = .81. In infants younger than 1 year, renal length was poorly estimated by a simple age-based formula.

CONCLUSION

Our proposed formula can be used to predict renal length in children older than 1 year.