Interpreting Risk and Impact Measures in Nursing Research: Implications for Evidence-based Practice.

Interpretation of “significant” research findings can be challenging. Outcome differences can be reported in relative or absolute terms. Belshe et al reported that children who received live attenuated influenza vaccine had a 3.9% risk of influenza during the subsequent season, while children who received inactivated influenza vaccine had an 8.6% risk. This difference in influenza rates can be expressed as the absolute risk reduction (ARR), also known as risk difference, which is simply the arithmetic difference between the outcome rates in the treatment and comparison groups (ARR = 8.6% – 3.9% = 4.7%).The same difference in rates can be expressed as the relative risk reduction (RRR), which is the ARR divided by the risk in the comparison group (RRR = 4.7% ÷ 8.6% = 0.55 or 55%).Relative measures, including RRR, relative risk, and odds ratios, are helpful in understanding the etiology of disease or the efficacy of interventions. The finding that live vaccine is associated with an influenza rate that is 55% lower than the rate following inactivated vaccine helps establish that live vaccine “works” better than inactivated vaccine – ie, that it actually protects better against influenza.Large relative effects, however, do not always translate into important clinical effects.The clinical or public health importance of disease-causing exposures or effective treatments may be better appreciated from absolute measures. The 55% relative decrease in influenza rates with live vaccine translates into a 4.7% absolute decrease (ie, among vaccinated children, 4.7% fewer will have influenza if live vaccine replaces inactivated vaccine).The ARR can be converted into a very useful quantity called the number needed to treat (NNT), which is the number of patients one would need to treat with an intervention in order to realize one improved outcome. The NNT is simply the reciprocal of the ARR, rounded up to the next whole number (NNT = 100% ÷ 4.7% = 21.3). Thus, one would have to vaccinate 22 children with live vaccine rather than inactivated vaccine in order to prevent one additional case of influenza.In an analogous way, the number needed to harm (NNH) compares adverse effects of different treatments. Belshe et al noted that among children less than 24 months of age, medically significant wheezing occurred within six weeks of the first vaccine dose in 3.2% of those receiving live vaccine compared with 2.0% of those receiving inactivated vaccine. Thus, the absolute risk increase or risk difference is 1.2%. The NNH can be calculated as 100% ÷ 1.2% = 83.3. Thus, for every 84 children in this age group vaccinated with live vaccine rather than inactivated vaccine, one additional child will have medically significant wheezing.By considering the absolute effect of an intervention and the NNT (along with the NNH for any significant adverse effects), one can make a better-informed decision as to whether a new treatment should be applied in practice.

See related article, pp 599–607 The high prevalence of hypertension means that even small refinements to treatment can have a significant impact on population health.1 However, hypertension treatment seems to have much more benefit for some patients than others, largely because of differences in baseline risk. Thus, 2 patients who achieve similar relative risk reduction can have much different absolute risk reductions based on their baseline level of risk. Table demonstrates this for stroke, using 4 studies that I admittedly chose to make this point.2–5 Age and baseline systolic blood pressure (BP) were the most obvious risk factors generating the widely disparate event rates in these studies. Current guidelines use these and other established risk factors to classify patients as at low, medium, or high risk, but with minimally different recommended approaches to treatment.1,6 Researchers have identified additional potential risk factors more recently, including variability in BP.7 View this table: Table. Progressive Increase in Absolute Risk Reduction With Increasing Baseline Risk and Stable Relative Risk Reduction in a Series of Clinical Trials In this issue of Hypertension , Juhanoja et al8 present research with several methodological strengths to argue that variability over several days in home BP measurements is associated with cardiovascular risk. Previous studies have demonstrated the association of adverse cardiovascular outcomes with variability among BP measurements separated by months and variability that occurs over the course of a single day.7 In particular, their study population combines 4 population-based, geographically diverse cohorts with excellent and long-term (mean 9.3 years) follow-up using validated outcome ascertainment methods. They are able to adjust for the baseline risk factors included in commonly used risk calculators (population averages had be used to estimate total and high-density lipoprotein cholesterol for ≈10% of …

Information on health risks conveyed in pamphlets, websites, and even medical journals often deceive health professionals and patients, either as a result of lack of understanding or as a result of the intention to manipulate opinions and decisions. This problem is even more meaningful when presenting the benefits using relative risk reduction (typically a large number) and, at the same time, showing the harms using absolute risk increase (usually a small number). Therefore, benefits are perceived as much higher, and damage is seen as much lower. In this study, one aims to evaluate the understanding and perceived information of health professionals and the general population concerning risk transmission in healthcare. The information was collected through a questionnaire that was made available online over a period of 3 months and 10 days. One hundred and fifty-four physicians (31.8%), 142 nurses (29.3%), and 189 people from the general population (39%) were considered. Of these, 58.4% were females, 40.2% males, and 1.4% did not specify. The majority of the respondents were relatively young (76.1% were 21 to 40 years old), and 87.3% of the general population held a higher degree. Less than half of physicians correctly recognized the relative risk and absolute risk reductions. Lower rates were seen for nurses and the general population. Notably, nurses do not seem to understand the information transmitted much better than the general population. This lack of statistical understanding of the risk variation concept among those involved in decision-making can impair the health of patients. It is important that information on health risks is corrected by replacing the presentation of relative risk variation by absolute risk variation, so that many can revise their opinions and decisions.

Abstract- Evidence-based medicine (EBM) has rapidly emerged as a new paradigm in medicine worldwide. The clinical medicine in twenty-first century could be the era of EBM. Randomized controlled trial has been regarded as the gold standard for evaluating the treatment effect of a new drug or a new therapy. The effect of a treatment versus controls may be expressed in relative or absolute measures. Relative measures include relative risk, relative risk reduction, and odds ratio. Absolute risk reduction and number needed to treat are absolute measures. For rational decision-making, absolute measures are more meaningful because they have taken baseline risk and the amount of clinical benefit into account. The number needed to treat (NNT), the reciprocal of the absolute risk reduction, is a useful estimate of treatment effect. Interpreting a NNT should be very cautious accompanied by information about the experimental treatment (including drugs and surgical procedures), the control treatment for comparison, the baseline risk of the study population, the length of the follow-up period, and an exact definition of the endpoint.

Understanding and expressing “Risk”

BackgroundAnalytic measurement of serum tumour markers is one of commonly used methods for cancer risk management in certain areas of the world (e.g. Taiwan). Recently, cancer screening based on multiple serum tumour markers has been frequently discussed. However, the risk–benefit outcomes appear to be unfavourable for patients because of the low sensitivity and specificity. In this study, cancer screening models based on multiple serum tumour markers were designed using machine learning methods, namely support vector machine (SVM), k-nearest neighbour (KNN), and logistic regression, to improve the screening performance for multiple cancers in a large asymptomatic population.MethodsAFP, CEA, CA19-9, CYFRA21-1, and SCC were determined for 20 696 eligible individuals. PSA was measured in men and CA15-3 and CA125 in women. A variable selection process was applied to select robust variables from these serum tumour markers to design cancer detection models. The sensitivity, specificity, positive predictive value (PPV), negative predictive value, area under the curve, and Youden index of the models based on single tumour markers, combined test, and machine learning methods were compared. Moreover, relative risk reduction, absolute risk reduction (ARR), and absolute risk increase (ARI) were evaluated.ResultsTo design cancer detection models using machine learning methods, CYFRA21-1 and SCC were selected for women, and all tumour markers were selected for men. SVM and KNN models significantly outperformed the single tumour markers and the combined test for men. All 3 studied machine learning methods outperformed single tumour markers and the combined test for women. For either men or women, the ARRs were between 0.003–0.008; the ARIs were between 0.119–0.306.ConclusionMachine learning methods outperformed the combined test in analysing multiple tumour markers for cancer detection. However, cancer screening based solely on the application of multiple tumour markers remains unfavourable because of the inadequate PPV, ARR, and ARI, even when machine learning methods were incorporated into the analysis.

Outcomes, Outcomes, Every where, nor any Stop to Think?

To understand the magnitude of risk of health events, such as cardiovascular diseases (CVD), related to poor oral health, both relative and absolute risk measures should be reported. Our aim was to investigate the extent to which absolute and relative measures of risk are reported in longitudinal studies that assess the association between oral health indicators (OHIs) and CVD. A systematic search of longitudinal studies investigating the association of any OHI with CVD was carried out using the Embase, Medline and Cochrane library databases. The search covered each database from its inception date to August 2021. Data about reporting relative and absolute risks of the relationship between CVD and OHI from the abstract were extracted. If the relative risk for OHIs and CVD was reported in the abstract, then the underlying absolute risks were searched from the full text, and it was assessed whether it was similarly adjusted for confounding than was the relative risk in the abstract. One hundred-six articles were included. From these, 85 (80%) studies reported the association of OHIs and CVD with one or more relative risks in the abstract. Of those 85studies, the underlying absolute risks were accessible or calculable from the abstract or full text of 60studies. However, of these 60studies, in only 10 (12%), the underlying absolute risks were similarly adjusted, as were the relative risks in the abstract. The absolute risks of CVD by OHIs were rarely reported without corresponding relative risks in the abstract (n=2, 2%). Median absolute risk difference in the CVD risk between exposure levels to which the first relative risk in the abstract referred was 1.8% (interquartile range 0.6-4.6, n=63). Focusing on relative risks over absolute risks was a common practice in literature. Reporting similarly adjusted underlying absolute risks of relative risks was rare in most studies, despite those being helpful for comprehending the magnitude of CVD-risk increase related to poor oral health. Current reporting practices could lead to an overinterpretation of risk increase of CVD related to poor oral health.

User's guide to the Chiropractic Literature-IB: how to use an article about therapy

Background and Aims Corticosteroids are an effective treatment for IgA nephropathy but are associated with considerable adverse events. The TESTING clinical trial showed an average treatment effect of a 47% relative risk reduction in the primary composite outcome for methylprednisolone versus placebo (hazard ratio 0.53, 95% CI 0.39–0.72). This is an aggregate result that cannot be applied to individual patients to make personalized treatment decisions, making it challenging to identify appropriate patients for corticosteroid treatment. To address this problem, we conducted a secondary analysis of the TESTING cohort to generate a model that can predict, for an individual patient, the probability that they will respond to methylprednisolone resulting in a lower risk of kidney disease progression. Method Time to the primary outcome (40% reduction in eGFR, kidney failure or death due to kidney disease) was first evaluated in a Cox proportional hazards model in which all potential treatment effect modifiers including demographic, clinical and MEST-C variables were evaluated as main effects using backwards elimination. The selected variables were then forced into a multivariable model along with treatment exposure and interaction terms between treatment and each other variable. This model was used to generate the predicted 4-year absolute risk of the primary outcome for each patient under separate counterfactual scenarios of being treated with methylprednisolone or placebo. The difference in risk between the two scenarios was the predicted individual treatment effect on absolute risk reduction (ARR). Model performance was assessed using discrimination plots, restricted mean survival time (RMST, an estimate of the additional time methylprednisolone provides without experiencing the primary outcome) and the C-statistic for benefit (ability of the model to discriminate between patients who got more versus less benefit from methylprednisolone). Results A total of 483 patients were included (median age 36 years, proteinuria 2.0 g/day, eGFR 57 mL/min). During 43 (median) months of follow-up, 176 participants experienced the primary outcome. Compared to the average ARR associated with methylprednisolone (16.1%, 95% CI 15.5–16.8), the predicted individual-level ARR was highly variable ranging from zero (for patients who experience minimal or no benefit) to more than 30% (for patients who experience considerable benefit) (Fig. 1, left panel). Patients with predicted ARR >10% had a substantially greater observed benefit from methylprednisolone (ARR 24%) compared to those with predicted ARR ≤10% (ARR −5%) (Fig. 1, right panel). A policy of treating patients with higher predicted benefit (ARR >10%) and not treating patients with low predicted benefit (ARR ≤10%) had a longer RMST than using random treatment allocation as was done in the main trial (1,194 v 1,028 days). The C-statistic for benefit was 0.63 (95% CI 0.56–0.70). Calibration plots showed considerable agreement between predicted and observed ARR. Findings were consistent in both the high-dose and reduced-dose methylprednisolone cohorts. The pattern of treatment effect modifiers was similar when the outcome was changed to annualized eGFR slope. Conclusion We have generated a model that can predict individual patient response to methylprednisolone and inform personalized treatment decisions in IgAN so that corticosteroid therapy can be targeted to those most likely to benefit.

Any medical innovation must first prove its benefits with reliable evidence from clinical trials. Evidence is commonly expressed using two metrics, summarizing treatment benefits based on either absolute risk reductions (ARRs) or relative risk reductions (RRRs). Both metrics are derived from the same data, but they implement conceptually distinct ideas. Here, we analyze these risk reductions measures from a causal modeling perspective. First, we show that ARR is equivalent to Δ P , while RRR is equivalent to causal power, thus clarifying the implicit causal assumptions. Second, we show how this formal equivalence establishes a relationship with causal Bayes nets theory, offering a basis for incorporating risk reduction metrics into a computational modeling framework. Leveraging these analyses, we demonstrate that under dynamically varying baseline risks, ARRs and RRRs lead to strongly diverging predictions. Specifically, the inherent assumption of a linear parameterization of the underlying causal graph can lead to incorrect conclusions when generalizing treatment benefits (e.g, predicting the effect of a vaccine in new populations with different baseline risks). Our analyses highlight the shared principles underlying risk reduction metrics and measures of causal strength, emphasizing the potential for explicating causal structure and inference in medical research. • Analyses explicate causal assumptions underlying different risk reduction metrics. • Absolute risk reduction = Δ P metric and relative risk reduction = causal power metric. • Equivalence of medical and causal metrics establishes a link to causal Bayes nets. • Tacit causal assumptions in risk metrics are crucial for generalizing medical effects. • Predictions of absolute and relative metrics diverge as baseline risks change.

The effect of raloxifene, a selective estrogen-receptor modulator, on coronary heart disease (CHD) and breast cancer is not established. We randomly assigned 10,101 postmenopausal women (mean age, 67.5 years) with CHD or multiple risk factors for CHD to 60 mg of raloxifene daily or placebo and followed them for a median of 5.6 years. The two primary outcomes were coronary events (i.e., death from coronary causes, myocardial infarction, or hospitalization for an acute coronary syndrome) and invasive breast cancer. As compared with placebo, raloxifene had no significant effect on the risk of primary coronary events (533 vs. 553 events; hazard ratio, 0.95; 95 percent confidence interval, 0.84 to 1.07), and it reduced the risk of invasive breast cancer (40 vs. 70 events; hazard ratio, 0.56; 95 percent confidence interval, 0.38 to 0.83; absolute risk reduction, 1.2 invasive breast cancers per 1000 women treated for one year); the benefit was primarily due to a reduced risk of estrogen-receptor-positive invasive breast cancers. There was no significant difference in the rates of death from any cause or total stroke according to group assignment, but raloxifene was associated with an increased risk of fatal stroke (59 vs. 39 events; hazard ratio, 1.49; 95 percent confidence interval, 1.00 to 2.24; absolute risk increase, 0.7 per 1000 woman-years) and venous thromboembolism (103 vs. 71 events; hazard ratio, 1.44; 95 percent confidence interval, 1.06 to 1.95; absolute risk increase, 1.2 per 1000 woman-years). Raloxifene reduced the risk of clinical vertebral fractures (64 vs. 97 events; hazard ratio, 0.65; 95 percent confidence interval, 0.47 to 0.89; absolute risk reduction, 1.3 per 1000). Raloxifene did not significantly affect the risk of CHD. The benefits of raloxifene in reducing the risks of invasive breast cancer and vertebral fracture should be weighed against the increased risks of venous thromboembolism and fatal stroke. (ClinicalTrials.gov number, NCT00190593 [ClinicalTrials.gov].).

BACKGROUND: Current use of hormonal contraception is associated with a 20-30% relative increase in the risk of breast cancer (BC) for women in the general population compared with never using. Longer duration of use is associated with higher risk, and the risk remains elevated above that of never users for at least 5 years after cessation. Most published data are for various formulations of the combined oral contraceptive pill, but associations are similar for progestogen-only contraceptives, including intrauterine devices. For women in the general population who use hormonal contraceptives in their 20s and 30s, when baseline BC risk for most women is low, these increased relative risks translate into only small increases in absolute risk. It is unclear whether use of hormonal contraceptives increases BC risk for women carrying a germline BRCA1 or BRCA2 pathogenic variant (PV). These women are at markedly higher risk of early-onset BC, so even slightly increased relative risks could translate to important increases in their absolute risk of BC. This study assessed the association between use of any hormonal contraception and BC risk for BRCA1 and BRCA2 PV carriers using individual participant data from four prospective cohorts. METHODS: Data from females born after 1920 with a PV in BRCA1 or BRCA2 and no history of cancer or bilateral mastectomy at cohort entry were analyzed. Cox regression models were used to estimate hazard ratios (HR) and 95% confidence intervals (CI) for BC (invasive disease or ductal carcinoma in situ) associated with use of hormonal contraceptives for at least 1 year, with age as the timescale, entry at cohort enrolment, and censoring at the earlier of bilateral mastectomy, death, diagnosis of another cancer or last follow-up. Analyses were adjusted for study, birth cohort, first-degree family history of BC, parity, premenopausal bilateral oophorectomy and menopausal status. Current use of hormonal contraceptives was defined as use within the previous year, to account for cessation of use due to BC symptoms or clinical investigation. RESULTS: Of 3,882 BRCA1 and 1,509 BRCA2 PV carriers, 53% and 71%, respectively had ever used hormonal contraceptives (for at least one year). The median cumulative duration of hormonal contraceptive use was 4.8 and 5.7 years, respectively. Overall, 488 BRCA1 and 191 BRCA2 PV carriers developed incident BC during a median of 5.9 and 5.6 years of follow-up, respectively. For BRCA1 PV carriers, use of hormonal contraceptives for at least 1 year was associated with increased BC risk (HR [95% CI]: 1.29 [1.04-1.60] p=0.019). BC risk increased with longer cumulative duration of hormonal contraceptive use (HR [95% CI]: 1.13 [0.88-1.45] p=0.35, 1.48 [1.11-1.96] p=0.007 and 1.56 [1.13-2.17] p=0.007 for 1-5, 6-10 and >10 years of use, respectively), with an estimated proportional increase in risk of 3% (1%-5%, p=0.002) for each additional year of use. For BRCA2 PV carriers, there was no evidence that current or past use, or cumulative duration of use, were associated with increased risk of BC, but confidence intervals on the HRs were wide. CONCLUSION: Hormonal contraceptive use is associated with an increased risk of BC for women carrying PVs in BRCA1 and risk increases with cumulative duration of use. Hormonal contraceptives are an important healthcare option for women; they provide excellent contraceptive efficacy and reduce risks of ovarian and endometrial cancer. Decisions about use of hormonal contraceptives in women at increased risk for BC due to BRCA1 PVs need to carefully weigh the risks and benefits; while shorter-term use may result in only small increases, prolonged cumulative use may result in larger increases in absolute BC risk that may not be acceptable to some women. Citation Format: Kelly-Anne Phillips, Joanne Kotsopoulos, Susan Domchek, James Chamberlain, Julie Bassett, Amber Aeilts, Irene Andrulis, Saundra Buys, Wanda Cui, Mary Daly, Andrea Eisen, William Foulkes, Michael Friedlander, Jacek Gronwald, John Hopper, Esther John, Beth Karlan, Raymond Kim, Jan Lubiński, Kelly Metcalfe, Katherine Nathanson, Christian F. Singer, Heather Symecko, Nadine Tung, Steven Narod, Mary Beth Terry, Roger Milne. Hormonal Contraception and Breast Cancer Risk for Carriers of Germline Pathogenic Variants in BRCA1 and BRCA2 [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PS10-01.

OPEN ACCESSJuly 15, 2013The Making Evidence-Based Medicine Simple Series — Therapy Module Michael Mojica, MD Michael Mojica, MD New York University School of Medicine Google Scholar More articles by this author https://doi.org/10.15766/mep_2374-8265.9478 SectionsAboutAbstract ToolsDownload Citations ShareFacebookTwitterEmail AbstractAbstract Introduction: The Making Evidence Based Medicine Simple Series is an evidence-based clinical practice curriculum. It consists of six modules that review the articles types addressing the most common clinical questions (i.e., diagnosis, therapy, harm, prognosis, meta-analysis, and clinical decision rules). Each module reviews an article from the medical literature using the Users Guide to Medical Literature criteria for validity, results, and applicability. Each modules includes a PowerPoint presentation with a script, a video of the presentation with narration, a quiz with answers, tips for teaching the module, a blank review form, and a completed critical article review for the article reviewed in the module. Methods: The Therapy Module reviews the appraisal of the validity, results, and applicability of a randomized clinical trial of steroid therapy for mild croup. The study addresses the following clinical question: In children with mild croup in the emergency department, will oral dexamethasone when compared to placebo result in a reduction in unscheduled return visits? In addition, this module reviews the following critical appraisal skills: (1) differentiating between a superiority, non-inferiority, and equivalence analysis, (2) identifying the role of randomization, allocation concealment, and blinding in reducing bias, (3) distinguishing between an intention to treat and per protocol analysis, (4) calculating absolute risk, absolute risk reduction, and relative risk from a 2 × 2 table, (5) distinguishing between clinical and statistical significance, and (6) calculating/interpreting a number needed to treat. Results: Evaluations by course participants have been extremely positive. Participants often comment that they have finally grasped important content that they have previously struggled with. The materials have been particularly useful to those who are leading a teaching session. Feedback from a 2013 course for pediatric emergency medicine fellows indicated that 92% of participants felt the Therapy Module was “extremely helpful.” In addition, 100% of participants agreed or strongly agreed that the modules improved their ability to appraise an article, and their ability to conduct evidence-based clinical practice. One-hundred percent also agreed or strongly agreed that the precourse materials were useful. Discussion: The Making Evidence Based Medicine Simple Series is a well-received evidence based clinical practice curriculum and the Therapy Module is an effective component of the series that can be used independently or with the other modules. Educational Objectives By the end of this session, learners will be able to: Evaluate the validity, results, and applicability of a therapy article.Develop a “clinical bottom line” for a therapy article.Describe the steps in a randomized controlled trial.Differentiate analysis types (i.e., superiority, non-inferiority, and equivalence studies).Outline a therapy question in population/patient, intervention/indicator, comparator/control, and outcome format.Understand the validity criteria for a therapy study.Define randomization, allocation concealment, and blinding.Identify the effects of randomization, allocation concealment, and blinding on bias.Distinguish between an intention to treat and per protocol analysis.Calculate absolute risk, absolute risk reduction, and relative risk from a 2 × 2 table.Calculate relative risk reduction and understand its limitation.Distinguish between clinical and statistical significance.Interpret statistical significance from a confidence interval of an absolute risk reduction and relative risk.Understand the applicability criteria for a therapy study.Calculate and interpret a number needed to treat and number needed to harm.Create a clinical bottom line for a therapy study. Sign up for the latest publications from MedEdPORTAL Add your email below FILES INCLUDEDReferencesRelatedDetails FILES INCLUDED Included in this publication: Critical Appraisal Form - Therapy - Blank.doc Critical Appraisal From - Dexamethasone for Mild Croup.pdf EBM Teaching Tips - General Approach to the Modules.pdf EBM Teaching Tips - Therapy.pdf Instructor's Guide - MESS Curriculum - Overview.pdf Instructor's Guide - MESS Curriculum - Therapy Module.pdf Lecture Notes - Therapy.pdf Lecture Slides - Therapy.pdf Lecture Slides - Therapy.ppt QUIZ ANSWERS Therapy.pdf QUIZ Therapy.pdf Video Screencast - Therapy.mp4 To view all publication components, extract (i.e., unzip) them from the downloaded .zip file. Download editor’s noteThis publication may contain technology or a display format that is no longer in use. Related The Making Evidence-Based Medicine Simple Series — Diagnostic Testing Module The Making Evidence-Based Medicine Simple Series — Therapy Module The Making Evidence-Based Medicine Simple Series — Harm Module The Making Evidence-Based Medicine Simple Series — Prognosis Module The Making Evidence-Based Medicine Simple Series — Meta-analysis Module The Making Evidence-Based Medicine Simple Series - Clinical Decision Rule Module Copyright & Permissions© 2013 Mojica. This is an open-access publication distributed under the terms of the Creative Commons Attribution-NonCommercial license.KeywordsEvidence-Based Clinical PracticeMeta-AnalysisJournal ClubDiagnostic TestingClinical Prediction RuleHarmPrognosisClinical Decision RuleTherapyEvidence-Based Medicine Disclosures None to report. Funding/Support None to report. Loading ...

BackgroundThe number of people diagnosed with stage I–III non-small cell lung cancer (NSCLC) is increasing, in part due to greater implementation of lung cancer screening and earlier detection. Definitive surgery, radiation, or chemoradiation are increasingly utilized along with adjunctive therapies that include chemotherapy, radiation, immune checkpoint inhibitors (ICIs), and receptor tyrosine kinase inhibitors (rTKIs). However, remedial and adverse effects exist for each modality that must be accounted for in individual treatment plans with curative intent. The objective of this study was to characterize the benefits and harms of curative intent therapy using a novel patient-centered precision approach.MethodsWe incorporated a precision medicine model to evaluate the benefits and harms using data from phase III randomized controlled trials (RCTs) or individual participant data meta-analyses of RCTs. We followed standard recommendations to assess benefit and harm with the absolute risk reduction (ARR) or absolute risk increase (ARI), and number needed-to-treat (NNT) for beneficial effect (NNTB) or NNT for harmful effect (NNTH). To measure the net effect of benefit and harm, we incorporated a novel summary statistic—the NNT for net effect (NNTnet), calculated as: 1/(ARR − ARI), or 1/(1/NNTB − 1/NNTH). We referenced guideline recommendations and interpreted results from the perspective of a hypothetical patient faced with choosing between treatment options; decision-making accounted for overall survival (OS) effects, what most patients have reported as acceptable mortality risk (≤2%) to gain 1 year of life, and guideline-endorsed treatment-associated mortality risk (≤5%).ResultsWe illustrated the NNTnet in screening and diagnosis. In definitive treatment, we identified: (I) overtreatment with lobectomy compared to segmentectomy in peripheral stage IA1–2 NSCLC (5-year OS: ARI, 3.2%; NNTH, 32); and (II) overtreatment with definitive tri-modality treatment for stage III NSCLC (i.e., induction chemoradiation followed by surgery), compared with concurrent chemoradiation without surgery, due to an excessively high 7–10% postoperative mortality with definitive tri-modality treatment and potential subsequent increased mortality within 1-year (two RCTs). In addition, we identified overtreatment with adjuvant radiation, compared to no adjuvant radiation, following complete resection of stage I–IIIB NSCLC (5-year OS: ARI, 5%; NNTB, 20) (14 RCTs). Furthermore, the harm of adjuvant radiation more than offsets the benefit of adjuvant chemotherapy (5-year OS: ARR, 4%; NNTB, 25): 1/(1/25 − 1/20), or −100. In other words, 100 patients treated with surgery and adjuvant radiation and chemotherapy, compared with surgery only, would result in one treatment-related death by 5 years. Finally, across four RCTs evaluating neoadjuvant chemo-ICI therapy, one in five participants with resectable IB–IIIA/B NSCLC did not subsequently receive curative surgery, resulting in potential undertreatment.ConclusionsThis study has important implications in clinical decision-making and the design of future trials to prevent overtreatment or undertreatment, maximize benefits, minimize harms, and achieve net benefit over harm in beneficent care for this growing population.