Chronic disease prevention and screening outcomes for patients with and without financial difficulty: a secondary analysis of the BETTER WISE cluster randomised controlled trial.

OBJECTIVE: Building on Existing Tools To improvE chronic disease pRevention and screening in primary care Wellness of cancer survIvorS and patiEnts (BETTER WISE) was designed to assess the effectiveness of a cancer and chronic disease prevention and screening (CCDPS) programme. Here, we compare outcomes in participants living with and without financial difficulty. DESIGN: Secondary analysis of a cluster-randomised controlled trial. SETTING: Patients of 59 physicians from 13 clinics enrolled between September 2018 and August 2019. PARTICIPANTS: 596 of 1005 trial participants who responded to a financial difficulty screening question at enrolment. INTERVENTION: 1-hour CCDPS visit versus usual care. OUTCOME MEASURES: Eligibility for a possible 24 CCDPS actions was assessed at baseline and the primary outcome was the percentage of eligible items that were completed at 12-month follow-up. We also compared the change in response to the financial difficulty screening question between baseline and follow-up. RESULTS: 55 of 265 participants (20.7%) in the control group and 69 of 331 participants (20.8%) in the intervention group reported living with financial difficulty. The primary outcome was 29% (95% CI 26% to 33%) for intervention and 23% (95% CI 21% to 26%) for control participants without financial difficulty (p=0.01). Intervention and control participants with financial difficulty scored 28% (95% CI 24% to 32%) and 32% (95% CI 27% to 38%), respectively (p=0.14). In participants who responded to the financial difficulty question at both time points (n=302), there was a net decrease in the percentage of participants who reported financial difficulty between baseline (21%) and follow-up (12%, p<0.001) which was similar in the control and intervention groups. The response rate to this question was only 51% at follow-up. CONCLUSION: The BETTER intervention improved uptake of CCDPS manoeuvres in participants without financial difficulty, but not in those living with financial difficulty. Improving CCDPS for people living with financial difficulty may require a different clinical approach or that social determinants be addressed concurrently with clinical and lifestyle needs or both. TRIAL REGISTRATION NUMBER: ISRCTN21333761.

Patient Experience With Resident Versus Staff Physicians: Results From a Cross-Sectional Patient Survey From Canadian Family Medicine Residencies.

BACKGROUND AND OBJECTIVES: We compared experiences of patients who reported usually being seen by a resident with those usually seen by a staff physician. METHODS: We analyzed responses to a patient experience survey distributed at 13 family medicine teaching practices affiliated with the University of Toronto between May and June 2020. We analyzed responses to seven questions pertaining to timely access, continuity, and patient-centeredness. We compared responses between two types of usual primary care clinicians and calculated odds ratios before and after adjustment for patient characteristics. RESULTS: We analyzed data from 6,545 unique surveys; 18.6% reported their usual clinician was a resident physician. Resident patients were more likely to be older, born outside of Canada, report a high school education or less, and report having difficulty making ends meet. Compared to patients of staff physicians, patients of resident physicians had lower odds of being able to see their preferred primary care clinician and lower odds of getting nonurgent care in a reasonable time. They also had lower odds of reporting patient-centered care, but we found no significant differences in whether the time for an urgent appointment was about right or whether accessing care after hours was easy. CONCLUSIONS: In our setting, patients who reported usually seeing resident physicians had worse continuity of care and timeliness for nonurgent care than patients who reported usually seeing staff physicians despite resident patients being older, sicker, and having a lower socioeconomic position. Postgraduate training programs need to test models to support access and continuity for resident patient panels.

Key informants perspectives on creating a high impact research department in family and community medicine: a qualitative project.

BACKGROUND: Primary care is integral to the health system and population health. Primary care research is still in development and most academic departments lack effective research investments. High impact primary care research programs are needed to advance the field to ensure a robust primary care system for the future. The project objective was to understand key informants' views of structures, functions, and processes required to create a high impact research program in an academic primary care department. METHODS: A descriptive qualitative project with key informants from research programs in primary care. Participants included international research leaders in primary care (n = 10), department of family and community researchers (n = 37) and staff (n = 9) in an academic primary care department, other university leaders (n = 3) and members of the departmental executive leadership team (1 department; 25 members). Semi-structured interviews (n = 27), and focus groups (n = 6) were audio recorded, transcribed, and analyzed using thematic analysis. We used a socioecological framework which described micro, meso, macro levels of influence. RESULTS: At the micro level despite barriers with respect to funding, protected time and lack of formal mentorship, personal motivation was a key factor. At the meso level, the organizational structure that promoted collaboration and a sense of connection emerged as a key factor. Specifically research leaders identified a research faculty development pipeline based on equity, diversity, inclusion, indigeneity, and accessibility principles with thematic areas of focus as key enablers. Lastly, at the macro level, an overarching culture and policies that promoted funding and primary care research was associated with high impact programs. CONCLUSION: The alignment/complementarity of micro, meso, and macro level factors influenced the creation of a high impact research department in primary care. High impact research in primary care is facilitated by the development of researchers through formalized and structured mentorship/sponsorship and a department culture that promote primary care research.

Cannabis-Induced Gastrointestinal Tract Symptoms in the Adult Population: A Systematic Review.

OBJECTIVE: Cannabinoid usage is widespread in the self-management of various medical ailments. However, adverse effects have been reported with use, especially pertaining to the gastrointestinal system in adults and aged patients. These range from nausea, vomiting, bloating, or abdominal pain. This systematic review of previously reported cannabis-induced gastrointestinal symptoms in the adult population from the literature provides an analysis of relevant data to enhance knowledge and awareness of this topic. METHODS: PubMed, Ovid MEDLINE, Cochrane Central, EMBASE, and Google Scholar databases were searched for relevant studies published from inception to March 2023. RESULTS: The search yielded 598 results, of which 13 were deemed relevant and underwent further review. These included two systematic reviews, one retrospective cohort study, one retrospective chart review, two cross-sectional studies, one survey, and six case reports. The Cochrane Risk Tool for bias analysis was applied where relevant. The total number of people in the studies selected for analysis was 79, 779. Twelve out of the thirteen included studies reported some type of gastrointestinal tract symptoms experienced in medical and/or recreational cannabis users ranging from nausea, vomiting, diarrhoea, abdominal pain to adult intussusception. CONCLUSION: Potential limitations include small sample sizes, variation in research methodologies, varied studied designs, and limited availability of data on specific populations such as geriatric users. Further research is warranted to add to current evidence pertaining to this emerging topic of significance, fill the broad knowledge gaps and contribute to evidence-based guidelines for healthcare professionals, ensuring safe prescribing practices and provision of quality care.

The Impact of a Point-of-Care Ultrasound (POCUS) Program to Diagnose and Manage Ascites in Home-based Palliative Care.

OBJECTIVE: The objective of this study was to examine the impact of a point-of-care ultrasound (POCUS) program among people with ascites receiving home-based palliative care by measuring the association of POCUS with ascites-related days spent out of the home, compared with outcomes before POCUS implementation. DESIGN: Open cohort study. SETTING AND PARTICIPANTS: Adults who had an ascites-related procedure (ARP) between January 1, 2014, and December 31, 2015 (ie, pre-POCUS) and January 1, 2019, and December 31, 2020 (ie, POCUS). An ARP was defined as using ultrasound to diagnose suspected ascites or a paracentesis with or without ultrasound guidance to manage ascites. METHODS: The main outcome was the rate of ascites-related days spent out of the home, defined as days out of the home for either ARPs or because of ascites-related hospital admissions, relative to the time patients were at risk for spending ascites-related days out of the home. RESULTS: Among the 103 pre-POCUS patients (mean age 68.0; 50.4% female), there were 161 ARPs with 12.4% occurring at home. Among the 127 POCUS patients (mean age 74.0; 52.0% female) there were 193 ARPs with 82.4% occurring at home. POCUS was associated with a significantly lower rate of ascites-related days spent outside the home (pre-POCUS rate of 33 days per 1000 patient days vs POCUS rate of 9 days; unadjusted incidence rate ratio (IRR), 3.86; 95% CI, 2.95-5.12; P < .001; adjusted IRR, 3.83; 95% CI, 1.27-11.54; P = .02). POCUS was also associated with a higher likelihood of ARPs occurring in the home [unadjusted odds ratio (OR), 32.44; 95% CI, 18.15-59.90; P < .001; adjusted OR, 48.99; 95% CI, 21.04-114.10; P < .001]. CONCLUSIONS AND IMPLICATIONS: The use of POCUS may contribute to maximizing time spent at home for palliative care patients with ascites. These findings support the use of POCUS for home-based palliative care programs.

Results from the BETTER WISE trial: a pragmatic cluster two arm parallel randomized controlled trial for primary prevention and screening in primary care during the COVID-19 pandemic.

BACKGROUND: Cancer and chronic diseases are a major cost to the healthcare system and multidisciplinary models with access to prevention and screening resources have demonstrated improvements in chronic disease management and prevention. Research demonstrated that a trained Prevention Practitioner (PP) in multidisciplinary team settings can improve achievement of patient level prevention and screening actions seven months after the intervention. METHODS: We tested the effectiveness of the PP intervention in a pragmatic two-arm cluster randomized controlled trial. Patients aged 40-65 were randomized at the physician level to an intervention group or to a wait-list control group. The intervention consisted of a patient visit with a PP. The PP received training in prevention and screening and use of the BETTER WISE tool kit. The effectiveness of the intervention was assessed using a composite outcome of the proportion of the eligible prevention and screening actions achieved between intervention and control groups at 12-months. RESULTS: Fifty-nine physicians were recruited in Alberta, Ontario, and Newfoundland and Labrador. Of the 1,005 patients enrolled, 733 (72.9%) completed the 12-month analysis. The COVID-19 pandemic occurred during the study time frame at which time nonessential prevention and screening services were not available and in-person visits with the PP were not allowed. Many patients and sites did not receive the intervention as planned. The mean composite score was not significantly higher in patients receiving the PP intervention as compared to the control group. To understand the impact of COVID on the project, we also considered a subset of patients who had received the intervention and who attended the 12-month follow-up visit before COVID-19. This assessment demonstrated the effectiveness of the BETTER visits, similar to the findings in previous BETTER studies. CONCLUSIONS: We did not observe an improvement in cancer and chronic disease prevention and screening (CCDPS) outcomes at 12 months after a BETTER WISE prevention visit: due to the COVID-19 pandemic, the study was not implemented as planned. Though benefits were described in those who received the intervention before COVID-19, the sample size was too small to make conclusions. This study may be a harbinger of a substantial decrease and delay in CCDPS activities under COVID restrictions. TRIAL REGISTRATION: ISRCTN21333761. Registered on 19/12/2016. http://www.isrctn.com/ISRCTN21333761 .

Assessment of Resident and AI Chatbot Performance on the University of Toronto Family Medicine Residency Progress Test: Comparative Study.

BACKGROUND: Large language model (LLM)-based chatbots are evolving at an unprecedented pace with the release of ChatGPT, specifically GPT-3.5, and its successor, GPT-4. Their capabilities in general-purpose tasks and language generation have advanced to the point of performing excellently on various educational examination benchmarks, including medical knowledge tests. Comparing the performance of these 2 LLM models to that of Family Medicine residents on a multiple-choice medical knowledge test can provide insights into their potential as medical education tools. OBJECTIVE: This study aimed to quantitatively and qualitatively compare the performance of GPT-3.5, GPT-4, and Family Medicine residents in a multiple-choice medical knowledge test appropriate for the level of a Family Medicine resident. METHODS: An official University of Toronto Department of Family and Community Medicine Progress Test consisting of multiple-choice questions was inputted into GPT-3.5 and GPT-4. The artificial intelligence chatbot's responses were manually reviewed to determine the selected answer, response length, response time, provision of a rationale for the outputted response, and the root cause of all incorrect responses (classified into arithmetic, logical, and information errors). The performance of the artificial intelligence chatbots were compared against a cohort of Family Medicine residents who concurrently attempted the test. RESULTS: GPT-4 performed significantly better compared to GPT-3.5 (difference 25.0%, 95% CI 16.3%-32.8%; McNemar test: P<.001); it correctly answered 89/108 (82.4%) questions, while GPT-3.5 answered 62/108 (57.4%) questions correctly. Further, GPT-4 scored higher across all 11 categories of Family Medicine knowledge. In 86.1% (n=93) of the responses, GPT-4 provided a rationale for why other multiple-choice options were not chosen compared to the 16.7% (n=18) achieved by GPT-3.5. Qualitatively, for both GPT-3.5 and GPT-4 responses, logical errors were the most common, while arithmetic errors were the least common. The average performance of Family Medicine residents was 56.9% (95% CI 56.2%-57.6%). The performance of GPT-3.5 was similar to that of the average Family Medicine resident (P=.16), while the performance of GPT-4 exceeded that of the top-performing Family Medicine resident (P<.001). CONCLUSIONS: GPT-4 significantly outperforms both GPT-3.5 and Family Medicine residents on a multiple-choice medical knowledge test designed for Family Medicine residents. GPT-4 provides a logical rationale for its response choice, ruling out other answer choices efficiently and with concise justification. Its high degree of accuracy and advanced reasoning capabilities facilitate its potential applications in medical education, including the creation of exam questions and scenarios as well as serving as a resource for medical knowledge or information on community services.

Predictors of Cognitive Change in Parkinson Disease: A 2-year Follow-up Study.

BACKGROUND: Mild cognitive impairment is common in Parkinson disease (PD-MCI). However, instability in this clinical diagnosis and variability in rates of progression to dementia raises questions regarding its utility for longitudinal tracking and prediction of cognitive change in PD. We examined baseline neuropsychological test and cognitive diagnosis predictors of cognitive change in PD. METHODS: Persons with PD, without dementia PD (N=138) underwent comprehensive neuropsychological assessment at baseline and were followed up to 2 years. Level II Movement Disorder Society criteria for PD-MCI and PD dementia (PDD) were applied annually. Composite global and domain cognitive z -scores were calculated based on a 10-test neuropsychological battery. RESULTS: Baseline diagnosis of PD-MCI was not associated with a change in global cognitive z -scores. Lower baseline attention and higher executive domain z -scores were associated with greater global cognitive z -score worsening regardless of cognitive diagnosis. Worse baseline domain z -scores in the attention and language domains were associated with progression to MCI or PDD, whereas higher baseline scores in all cognitive domains except executive function were associated with clinical and psychometric reversion to "normal" cognition. CONCLUSIONS: Lower scores on cognitive tests of attention were predictive of worse global cognition over 2 years of follow-up in PD, and lower baseline attention and language scores were associated with progression to MCI or PDD. However, PD-MCI diagnosis per se was not predictive of cognitive decline over 2 years. The association between higher executive domain z -scores and greater global cognitive worsening is probably a spurious result.

Evaluating the Applicability of Existing Lexicon-Based Sentiment Analysis Techniques on Family Medicine Resident Feedback Field Notes: Retrospective Cohort Study.

BACKGROUND: Field notes, a form for resident-preceptor clinical encounter feedback, are widely adopted across Canadian medical residency training programs for documenting residents' performance. This process generates a sizeable cumulative collection of feedback text, which is difficult for medical education faculty to navigate. As sentiment analysis is a subfield of text mining that can efficiently synthesize the polarity of a text collection, sentiment analysis may serve as an innovative solution. OBJECTIVE: This study aimed to examine the feasibility and utility of sentiment analysis using 3 popular sentiment lexicons on medical resident field notes. METHODS: We used a retrospective cohort design, curating text data from University of Toronto medical resident field notes gathered over 2 years (from July 2019 to June 2021). Lexicon-based sentiment analysis was applied using 3 standardized dictionaries, modified by removing ambiguous words as determined by a medical subject matter expert. Our modified lexicons assigned words from the text data a sentiment score, and we aggregated the word-level scores to a document-level polarity score. Agreement between dictionaries was assessed, and the document-level polarity was correlated with the overall preceptor rating of the clinical encounter under assessment. RESULTS: Across the 3 original dictionaries, approximately a third of labeled words in our field note corpus were deemed ambiguous and were removed to create modified dictionaries. Across the 3 modified dictionaries, the mean sentiment for the "Strengths" section of the field notes was mildly positive, while it was slightly less positive in the "Areas of Improvement" section. We observed reasonable agreement between dictionaries for sentiment scores in both field note sections. Overall, the proportion of positively labeled documents increased with the overall preceptor rating, and the proportion of negatively labeled documents decreased with the overall preceptor rating. CONCLUSIONS: Applying sentiment analysis to systematically analyze field notes is feasible. However, the applicability of existing lexicons is limited in the medical setting, even after the removal of ambiguous words. Limited applicability warrants the need to generate new dictionaries specific to the medical education context. Additionally, aspect-based sentiment analysis may be applied to navigate the more nuanced structure of texts when identifying sentiments. Ultimately, this will allow for more robust inferences to discover opportunities for improving resident teaching curriculums.

Quality indices for topic model selection and evaluation: a literature review and case study.

BACKGROUND: Topic models are a class of unsupervised machine learning models, which facilitate summarization, browsing and retrieval from large unstructured document collections. This study reviews several methods for assessing the quality of unsupervised topic models estimated using non-negative matrix factorization. Techniques for topic model validation have been developed across disparate fields. We synthesize this literature, discuss the advantages and disadvantages of different techniques for topic model validation, and illustrate their usefulness for guiding model selection on a large clinical text corpus. DESIGN, SETTING AND DATA: Using a retrospective cohort design, we curated a text corpus containing 382,666 clinical notes collected between 01/01/2017 through 12/31/2020 from primary care electronic medical records in Toronto Canada. METHODS: Several topic model quality metrics have been proposed to assess different aspects of model fit. We explored the following metrics: reconstruction error, topic coherence, rank biased overlap, Kendall's weighted tau, partition coefficient, partition entropy and the Xie-Beni statistic. Depending on context, cross-validation and/or bootstrap stability analysis were used to estimate these metrics on our corpus. RESULTS: Cross-validated reconstruction error favored large topic models (K ≥ 100 topics) on our corpus. Stability analysis using topic coherence and the Xie-Beni statistic also favored large models (K = 100 topics). Rank biased overlap and Kendall's weighted tau favored small models (K = 5 topics). Few model evaluation metrics suggested mid-sized topic models (25 ≤ K ≤ 75) as being optimal. However, human judgement suggested that mid-sized topic models produced expressive low-dimensional summarizations of the corpus. CONCLUSIONS: Topic model quality indices are transparent quantitative tools for guiding model selection and evaluation. Our empirical illustration demonstrated that different topic model quality indices favor models of different complexity; and may not select models aligning with human judgment. This suggests that different metrics capture different aspects of model goodness of fit. A combination of topic model quality indices, coupled with human validation, may be useful in appraising unsupervised topic models.

Opportunities to improve quality of care for cancer survivors in primary care: findings from the BETTER WISE study.

PURPOSE: The BETTER WISE (Building on Existing Tools to Improve Chronic Disease Prevention and Screening in Primary Care for Wellness of Cancer Survivors and Patients) intervention is an evidence-based approach to prevention and screening for cancers and chronic diseases in primary care that also includes comprehensive follow-up for breast, prostate and colorectal cancer survivors. We describe the process of harmonizing cancer survivorship guidelines to create a BETTER WISE cancer surveillance algorithm and describe both the quantitative and qualitative findings for BETTER WISE participants who were breast, prostate or colorectal cancer survivors. We describe the results in the context of the COVID-19 pandemic. METHODS: We reviewed high-quality survivorship guidelines to create a cancer surveillance algorithm. We conducted a cluster randomized trial in three Canadian provinces with two composite index outcome measured 12 months after baseline, and also collected qualitative feedback on the intervention. RESULTS: There were 80 cancer survivors for whom we had baseline and follow-up data. Differences between the composite indices in the two study arms were not statistically significant, although a post hoc analysis suggested the COVID-19 pandemic was a key factor in these results. Qualitative finding suggested that participants and stakeholders generally viewed BETTER WISE positively and emphasized the effects of the pandemic. CONCLUSIONS AND IMPLICATIONS FOR CANCER SURVIVORS: BETTER WISE shows promise for providing an evidence-based, patient-centred, comprehensive approach to prevention, screening and cancer surveillance for cancer survivors in the primary care setting. TRIAL REGISTRATION: ISRCTN21333761. Registered on December 19, 2016, http://www.isrctn.com/ISRCTN21333761 .

Emulating a Target Trial Using Primary-Care Electronic Health Records: Sodium-Glucose Cotransporter 2 Inhibitor Medications and Hemoglobin A1c.

Substantial effort has been dedicated to conducting randomized controlled experiments to generate clinical evidence for diabetes treatment. Randomized controlled experiments are the gold standard for establishing cause and effect. However, due to their high cost and time commitment, large observational databases such as those comprised of electronic health record (EHR) data collected in routine primary care may provide an alternative source with which to address such causal objectives. We used a Canadian primary-care data repository housed at the University of Toronto (Toronto, Ontario, Canada) to emulate a randomized experiment. We estimated the effectiveness of sodium-glucose cotransporter 2 inhibitor (SGLT-2i) medications for patients with diabetes using hemoglobin A1c (HbA1c) as a primary outcome and marker for glycemic control from 2018 to 2021. We assumed an intention-to-treat analysis for prescribed treatment, with analyses based on the treatment assigned rather than the treatment eventually received. We defined the causal contrast of interest as the net change in HbA1c (percent) between the group receiving the standard of care versus the group receiving SGLT-2i medication. Using a counterfactual framework, marginal structural models demonstrated a reduction in mean HbA1c level with the initiation of SGLT-2i medications. These findings provided effect sizes similar to those from earlier clinical trials on assessing the effectiveness of SGLT-2i medications.

Using ICD-9 diagnostic codes for external validation of topic models derived from primary care electronic medical record clinical text data.

Background/Objectives: Unsupervised topic models are often used to facilitate improved understanding of large unstructured clinical text datasets. In this study we investigated how ICD-9 diagnostic codes, collected alongside clinical text data, could be used to establish concurrent-, convergent- and discriminant-validity of learned topic models. Design/Setting: Retrospective open cohort design. Data were collected from primary care clinics located in Toronto, Canada between 01/01/2017 through 12/31/2020. Methods: We fit a non-negative matrix factorization topic model, with K = 50 latent topics/themes, to our input document term matrix (DTM). We estimated the magnitude of association between each Boolean-valued ICD-9 diagnostic code and each continuous latent topical vector. We identified ICD-9 diagnostic codes most strongly associated with each latent topical vector; and qualitatively interpreted how these codes could be used for external validation of the learned topic model. Results: The DTM consisted of 382,666 documents and 2210 words/tokens. We correlated concurrently assigned ICD-9 diagnostic codes with learned topical vectors, and observed semantic agreement for a subset of latent constructs (e.g. conditions of the breast, disorders of the female genital tract, respiratory disease, viral infection, eye/ear/nose/throat conditions, conditions of the urinary system, and dermatological conditions, etc.). Conclusions: When fitting topic models to clinical text corpora, researchers can leverage contemporaneously collected electronic medical record data to investigate the external validity of fitted latent variable models.

Comparison of Methods for Estimating Temporal Topic Models From Primary Care Clinical Text Data: Retrospective Closed Cohort Study.

BACKGROUND: Health care organizations are collecting increasing volumes of clinical text data. Topic models are a class of unsupervised machine learning algorithms for discovering latent thematic patterns in these large unstructured document collections. OBJECTIVE: We aimed to comparatively evaluate several methods for estimating temporal topic models using clinical notes obtained from primary care electronic medical records from Ontario, Canada. METHODS: We used a retrospective closed cohort design. The study spanned from January 01, 2011, through December 31, 2015, discretized into 20 quarterly periods. Patients were included in the study if they generated at least 1 primary care clinical note in each of the 20 quarterly periods. These patients represented a unique cohort of individuals engaging in high-frequency use of the primary care system. The following temporal topic modeling algorithms were fitted to the clinical note corpus: nonnegative matrix factorization, latent Dirichlet allocation, the structural topic model, and the BERTopic model. RESULTS: Temporal topic models consistently identified latent topical patterns in the clinical note corpus. The learned topical bases identified meaningful activities conducted by the primary health care system. Latent topics displaying near-constant temporal dynamics were consistently estimated across models (eg, pain, hypertension, diabetes, sleep, mood, anxiety, and depression). Several topics displayed predictable seasonal patterns over the study period (eg, respiratory disease and influenza immunization programs). CONCLUSIONS: Nonnegative matrix factorization, latent Dirichlet allocation, structural topic model, and BERTopic are based on different underlying statistical frameworks (eg, linear algebra and optimization, Bayesian graphical models, and neural embeddings), require tuning unique hyperparameters (optimizers, priors, etc), and have distinct computational requirements (data structures, computational hardware, etc). Despite the heterogeneity in statistical methodology, the learned latent topical summarizations and their temporal evolution over the study period were consistently estimated. Temporal topic models represent an interesting class of models for characterizing and monitoring the primary health care system.

Association Between Clinic-Reported Third Next Available Appointment and Patient-Reported Access to Primary Care.

This cross-sectional study examines the association of the third next available appointment with patient-reported measures of access in primary care settings.

Examining Access to Primary Care for People With Opioid Use Disorder in Ontario, Canada: A Randomized Clinical Trial.

Importance: People with opioid use disorder are less likely than others to have a primary care physician. Objective: To determine if family physicians are less likely to accept people with opioid use disorder as new patients than people with diabetes. Design, Setting, and Participants: This randomized clinical trial used an audit design to survey new patient intake at randomly selected family physicians in Ontario, Canada. Eligible physicians were independent practitioners allowed to prescribe opioids who were located in an office within 50 km of a population center greater than 20 000 people. A patient actor made unannounced telephone calls to family physicians asking for a new patient appointment. The data were analyzed in September 2021. Intervention: In the first randomly assigned scenario, the patient actor played a role of patient with diabetes in treatment with an endocrinologist. In the second scenario, the patient actor played a role of a patient with opioid use disorder undergoing methadone treatment with an addiction physician. Main Outcomes and Measures: Total offers of a new patient appointment; a secondary analysis compared the proportions of patients offered an appointment stratified by gender, population, model of care, and years in practice. Results: Of a total 383 family physicians included in analysis, a greater proportion offered a new patient appointment to a patient with diabetes (21 of 185 physicians [11.4%]) than with opioid use disorder (8 of 198 physicians [4.0%]) (absolute difference, 7.4%; 95% CI, 2.0 to 12.6; P = .007). Physicians with more than 20 years in practice were almost 13 times less likely to offer an appointment to a patient with opioid use disorder compared with diabetes (1 of 108 physicians [0.9%] vs 10 of 84 physicians [11.9%]; absolute difference, 11.0; 95% CI, 3.8 to 18.1; P = .001). Women were almost 5 times less likely (3 of 111 physicians [2.7%] vs 14 of 114 physicians [12.3%]; absolute difference, 9.6%; 95% CI, 2.4 to 16.3; P = .007) to offer an appointment to a patient with opioid use disorder than with diabetes. Conclusions and Relevance: In this randomized clinical trial, family physicians were less likely to offer a new patient appointment to a patient with opioid use disorder compared with a patient with diabetes. Potential health system solutions to this disparity include strengthening policies for accepting new patients, improved compensation, and clinician anti-oppression training. Trial Registration: ClinicalTrials.gov Identifier: NCT05484609.

Precarious work among personal support workers in the Greater Toronto Area: a respondent-driven sampling study.

BACKGROUND: The COVID-19 pandemic has highlighted the role of personal support workers (PSWs) in health care, as well as their work conditions. Our study aimed to understand the characteristics of the PSW workforce, their work conditions and their job security, as well as to explore the health of PSWs and the impact of precarious employment on their health. METHODS: Our community-based participatory action research focused on PSWs in the Greater Toronto Area. We administered an online, cross-sectional survey between June and December 2020 using respondent-driven sampling. Data on sociodemographics, employment precarity, worker empowerment and health status were collected. We assessed the association between precarious employment and health using multivariable logistic regression models. RESULTS: We contacted 739 PSWs, and 664 consented to participate. Overall, 658 (99.1%) completed at least part of the survey. Using data adjusted for our sampling approach, the participants were predominantly Black (76.5%, 95% confidence interval [CI] 68.2%-84.9%), women (90.1%, 95% CI 85.1%-95.1%) and born outside of Canada (97.4%, 95% CI 94.9%-99.9%). Most worked in home care (43.9%, 95% CI 35.2%-52.5%) or long-term care (34.5%, 95% CI 27.4%-42.0%). Although most participants had at least some postsecondary education (unadjusted proportion = 83.4%, n = 529), more than half were considered low income (55.1%, 95% CI 46.3%-63.9%). Most participants were precariously employed (86.5%, 95% CI 80.7%-92.4%) and lacked paid sick days (89.5%, 95% CI 85.8%-93.3%) or extended health benefits (74.1%, 95% CI 66.8%-81.4%). Nearly half of the participants described their health as less than very good (46.7%, 95% CI 37.9%-55.5%). Employment precarity was significantly associated with higher risk of depression (odds ratio 1.02, 95% CI 1.01-1.03). INTERPRETATION: Despite being key members of health care teams, most PSWs were precariously employed with low wages that keep them in poverty; the poor work conditions they faced could be detrimental to their physical and mental health. Equitable strategies are needed to provide decent work conditions for PSWs and to improve their health.

The working conditions for personal support workers in the Greater Toronto Area during the COVID-19 pandemic: a mixed-methods study.

OBJECTIVE: During the height of the COVID-19 pandemic, personal support workers (PSWs) were heralded as healthcare 'heroes' as many of them cared for high-risk, vulnerable older populations, and worked in long-term care, which experienced a high number of COVID-19 outbreaks and deaths. While essential to the healthcare workforce, there is little understanding of PSW working conditions during the pandemic. The aim of our study was to examine the working conditions (including job security, work policies, and personal experiences) for PSWs in the Greater Toronto Area during the COVID-19 pandemic from the perspectives of PSWs. METHODS: This study used a mixed-methods design. From June to December 2020, we conducted a survey of 634 PSWs to understand their working conditions during the COVID-19 pandemic. Semi-structured interviews with 31 survey respondents were conducted from February to May 2021 to understand in greater depth how working conditions were impacting the well-being of PSWs. RESULTS: We found PSWs faced a range of challenges related to COVID-19, including anxiety about contracting COVID-19, reduced work hours, taking leaves of absences, concerns about job security, and losing childcare. While the COVID-19 pandemic highlighted the PSW workforce and their importance to the healthcare system (especially in the long-term care system), pre-existing poor work conditions of insecure jobs with no paid sick days and benefits exacerbated COVID-19-related challenges. Despite these hardships, PSWs were able to rely on their mental resilience and passion for their profession to cope with challenges. CONCLUSION: Significant changes need to be made to improve PSW working conditions. Better compensation, increased job security, decreased workload burden, and mental health supports are needed.

RéSUMé: OBJECTIF: Au plus fort de la pandémie de COVID-19, les préposés aux bénéficiaires (PAB) étaient salués comme les « héros ¼ des soins de santé, car beaucoup s'occupaient de populations vulnérables et âgées à risque élevé et travaillaient dans le milieu des soins de longue durée, qui était aux prises avec un grand nombre d'éclosions de COVID-19 et de décès. Malgré l'importance névralgique des PAB dans la main-d'œuvre des soins de santé, leurs conditions de travail durant la pandémie sont peu connues. Nous avons cherché à examiner les conditions de travail (sécurité d'emploi, régimes de travail et expériences personnelles) des PAB dans la région du Grand Toronto durant la pandémie de COVID-19 selon la perspective de ces PAB. MéTHODE: Nous avons utilisé un plan d'étude à méthodes mixtes. De juin à décembre 2020, nous avons sondé 634 PAB pour connaître leurs conditions de travail durant la pandémie de COVID-19. Des entretiens semi-directifs ont été menés auprès de 31 répondants et répondantes entre les mois de février et de mai 2021 pour approfondir notre compréhension de l'impact des conditions de travail des PAB sur leur bien-être. RéSULTATS: Nous avons constaté que les PAB ont connu de nombreuses difficultés en lien avec la COVID-19, dont l'anxiété à l'idée de transmettre la COVID-19, les heures de travail réduites, la prise de congés, les craintes pour leur sécurité d'emploi et la perte de services de garde. Bien que la pandémie de COVID-19 ait braqué les projecteurs sur les PAB et leur importance pour le système de soins de santé (surtout dans le système de soins de longue durée), leurs mauvaises conditions de travail préexistantes ­ des emplois précaires sans congés de maladie payés ni avantages sociaux ­ ont exacerbé les problèmes posés par la COVID-19. Malgré ces épreuves, les PAB ont pu compter sur leur résilience mentale et leur passion pour la profession pour faire face aux difficultés. CONCLUSION: D'importants changements doivent être apportés pour améliorer les conditions de travail des PAB. De meilleurs salaires, une plus grande sécurité d'emploi, une charge de travail allégée et des mesures d'appui à la santé mentale sont nécessaires.

Sociodemographic differences in patient experience with primary care during COVID-19: results from a cross-sectional survey in Ontario, Canada.

PURPOSE: We sought to understand patients' care-seeking behaviours early in the pandemic, their use and views of different virtual care modalities, and whether these differed by sociodemographic factors. METHODS: We conducted a multisite cross-sectional patient experience survey at 13 academic primary care teaching practices between May and June 2020. An anonymised link to an electronic survey was sent to a subset of patients with a valid email address on file; sampling was based on birth month. For each question, the proportion of respondents who selected each response was calculated, followed by a comparison by sociodemographic characteristics using χ2 tests. RESULTS: In total, 7532 participants responded to the survey. Most received care from their primary care clinic during the pandemic (67.7%, 5068/7482), the majority via phone (82.5%, 4195/5086). Among those who received care, 30.53% (1509/4943) stated that they delayed seeking care because of the pandemic. Most participants reported a high degree of comfort with phone (92.4%, 3824/4139), video (95.2%, 238/250) and email or messaging (91.3%, 794/870). However, those reporting difficulty making ends meet, poor or fair health and arriving in Canada in the last 10 years reported lower levels of comfort with virtual care and fewer wanted their practice to continue offering virtual options after the pandemic. CONCLUSIONS: Our study suggests that newcomers, people living with a lower income and those reporting poor or fair health have a stronger preference and comfort for in-person primary care. Further research should explore potential barriers to virtual care and how these could be addressed.

Marginal Structural Models Using Calibrated Weights With SuperLearner: Application to Type II Diabetes Cohort.

As different scientific disciplines begin to converge on machine learning for causal inference, we demonstrate the application of machine learning algorithms in the context of longitudinal causal estimation using electronic health records. Our aim is to formulate a marginal structural model for estimating diabetes care provisions in which we envisioned hypothetical (i.e. counterfactual) dynamic treatment regimes using a combination of drug therapies to manage diabetes: metformin, sulfonylurea and SGLT-2i. The binary outcome of diabetes care provisions was defined using a composite measure of chronic disease prevention and screening elements [27] including (i) primary care visit, (ii) blood pressure, (iii) weight, (iv) hemoglobin A1c, (v) lipid, (vi) ACR, (vii) eGFR and (viii) statin medication. We used several statistical learning algorithms to describe causal relationships between the prescription of three common classes of diabetes medications and quality of diabetes care using the electronic health records contained in National Diabetes Repository. In particular, we generated an ensemble of statistical learning algorithms using the SuperLearner framework based on the following base learners: (i) least absolute shrinkage and selection operator, (ii) ridge regression, (iii) elastic net, (iv) random forest, (v) gradient boosting machines, and (vi) neural network. Each statistical learning algorithm was fitted using the pseudo-population generated from the marginalization of the time-dependent confounding process. Covariate balance was assessed using the longitudinal (i.e. cumulative-time product) stabilized weights with calibrated restrictions. Our results indicated that the treatment drop-in cohorts (with respect to metformin, sulfonylurea and SGLT-2i) may have improved diabetes care provisions in relation to treatment naïve (i.e. no treatment) cohort. As a clinical utility, we hope that this article will facilitate discussions around the prevention of adverse chronic outcomes associated with type II diabetes through the improvement of diabetes care provisions in primary care.