HUMBEHV 2AP3 - Vision

Principles of Applied Psychology: Children's Vision Screening

The Critical Need for Early Visual Experience

  • Low Eye Exam Rates: In Ontario, only 6\% of children aged 0-6 years had an eye exam in 2013. This highlights a significant gap in early vision care.

  • Brain Development: Early visual experience is absolutely crucial for healthy brain development. Undetected and untreated vision problems can have long-lasting developmental impacts.

Understanding Amblyopia

  • Definition: Amblyopia, commonly known as "lazy eye," refers to reduced cortical vision in an otherwise sound eye. This occurs due to abnormal visual experience early in life, leading to the brain favoring the stronger eye and neglecting signals from the weaker eye.

  • Causes:

    • Unequal Refractive Errors (Anisometropia): This involves significant differences in refractive power between the two eyes. For example, one eye might be near-sighted while the other is far-sighted.

    • Misalignment (Strabismus): This condition involves weaker eye muscles in one eye, causing it to turn inward, outward, upward, or downward. The misaligned eye does not foveate on the same target as the other eye, leading to the brain suppressing the image from the misaligned eye.

  • Prevalence: Amblyopia affects a notable 2-8\% of kindergarten children, making it a significant public health concern for this age group.

  • Treatment Efficacy: Treatment for amblyopia is most effective when initiated before the age of 7 years. This emphasizes the critical window for early detection and intervention.

Evaluating School-Based Vision Screening Programs

  • Critical Evaluation: The presentation focuses on critically evaluating the efficacy of school-based vision screening programs, particularly referencing the research by Nishimura et al. (2018) published in BMJ Open.

  • Screening Process (Nishimura et al., 2018):

    • Five Tests: Children undergo a series of five vision screening tests.

    • Outcome: Based on the test results, children either "Pass" or are "Referred."

    • Follow-up: Referred children are scheduled for an optometry exam, often conducted with parent presence and requiring specialized eye drops for a comprehensive assessment.

    • Treatment: If vision problems are detected, free glasses can be provided to address the issues.

  • Specific Screening Tests Used:

    1. Cambridge Crowded Acuity Cards: Measures visual acuity in a format suitable for young children.

    2. Preschool Randot Stereo Acuity: Assesses depth perception, which is often affected in amblyopia.

    3. Plusoptix Autorefractor: An automated device that quickly measures refractive errors in both eyes.

    4. Spot Autorefractor: Another type of automated refractometer, similar in function to the Plusoptix.

    5. Pediatric Vision Scanner (PVS): A device designed to screen for various vision problems in children.

  • Passing Criteria: To "Pass" the screening, a child must successfully pass all five of the aforementioned tests.

Key Concepts in Screening: Sensitivity and Specificity

This section delves into fundamental epidemiological concepts used to evaluate the performance of diagnostic and screening tests.

  • Four Possible Test Outcomes:

    • True Positive (Hit): The screening test correctly identifies an individual who has the condition (e.g., a sick person correctly identified as sick).

    • False Positive: The screening test incorrectly identifies an individual who does not have the condition (e.g., a healthy person incorrectly identified as sick, leading to a "false alarm").

    • Miss (False Negative): The screening test incorrectly identifies an individual who does not have the condition, when in reality they do have it (e.g., a sick person incorrectly identified as healthy, leading to a "missed problem").

    • Correct Rejection (True Negative): The screening test correctly identifies an individual who does not have the condition (e.g., a healthy person correctly identified as healthy).

  • Sensitivity: This metric quantifies the test's ability to correctly identify those with the condition.

    • Formal Definition: The probability of a positive test result, given that the individual is positive for the condition.

    • Formula: ext{Sensitivity} = \frac{\text{True Positives}}{\text{True Positives} + \text{False Negatives}}

    • Analogy: How many relevant sick people are correctly identified as having the condition?

  • Specificity: This metric quantifies the test's ability to correctly identify those without the condition.

    • Formal Definition: The probability of a negative test result, given that the individual is negative for the condition.

    • Formula: ext{Specificity} = \frac{\text{True Negatives}}{\text{True Negatives} + \text{False Positives}}

    • Analogy: How many negative selected elements (healthy people) are truly negative (identified as not having the condition)?

Screening Program Efficacy: Nishimura et al. (2018) Results

Using data from a school-based vision screening program, the following results were observed for a total of 712 children across Junior and Senior Kindergarten:

  • Overall Statistics:

    • Total with Eye Problem: 180

    • Total without Eye Problem: 532

    • Referred by Screening: 427 (Includes 151 True Positives and 276 False Positives)

    • Passed Screening: 285 (Includes 29 False Negatives and 256 True Negatives)

    • Overall Sensitivity: 0.84 (95% CI: 0.78-0.89)

    • Overall Specificity: 0.48 (95% CI: 0.44-0.53) - Note the relatively low specificity, indicating many false alarms.

  • Junior Kindergarten (Ages 4 to 5 years, n = 442):

    • Sensitivity: 0.80

    • Specificity: 0.42

    • Breakdown of Outcomes:

      • Missed problems (False Negatives): 5\% of children

      • False alarms (False Positives): 43\% of children

      • Caught problems (True Positives): 21\% of children

      • Passed screening results (correctly negative): 36\% of children

      • Failed screening: 64\% of children

      • Passed screening: 36\% of children

  • Senior Kindergarten (Ages 5 to 6 years, n = 267):

    • Sensitivity: 0.91

    • Specificity: 0.58

    • Breakdown of Outcomes:

      • Missed problems (False Negatives): 2\% of children

      • False alarms (False Positives): 32\% of children

      • Caught problems (True Positives): 22\% of children

      • Passed screening results (correctly negative): 46\% of children

      • Failed screening: 54\% of children

      • Passed screening: 46\% of children

Receiver Operating Characteristic (ROC) Curves and AUC

  • ROC Curve: A ROC curve is a graphical plot that illustrates the diagnostic ability of a binary classifier system as its discrimination threshold is varied.

    • It plots the True Positive Rate (Sensitivity) on the Y-axis against the False Positive Rate (1 - ext{Specificity}) on the X-axis.

    • Perfect Classifier: A perfect classifier would have a point at (0, 1), meaning 100\% sensitivity and 100\% specificity (0\% false positive rate).

    • Random Classifier: A random classifier performs no better than chance and is represented by a diagonal line from (0,0) to (1,1).

    • Interpretation: Classifiers performing better than random will have a curve that bows towards the top-left corner of the graph.

  • Area Under the Curve (AUC): The AUC quantifies the overall ability of the test to discriminate between those with the condition and those without it.

    • An AUC of 1.0 indicates a perfect classifier.

    • An AUC of 0.5 indicates a classifier that performs no better than random chance.

    • Higher AUC values suggest a more accurate screening tool.

  • AUC Values for Different Screening Tools (Nishimura et al. (2018)):

    • Acuity (Any problem, n=691): AUC = 0.685 (p<0.001)

    • Plusoptix-cylinder (Any problem, n=672): AUC = 0.822 (p<0.001)

    • Plusoptix - SE (Any problem, n=672): AUC = 0.507 (p=0.80) - Indicates performance close to random, suggesting this particular measure from Plusoptix is not effective for detecting any problem.

    • Spot - cylinder (Any problem, n=707): AUC = 0.836 (p<0.001)

    • Spot - SE (Any problem, n=707): AUC = 0.537 (p=0.13) - Similar to Plusoptix - SE, poor performance for detecting any problem.

    • Randot (Any problem, n=694): AUC = 0.615 (p<0.001)

    • PVS (Any problem, n=699): AUC = 0.610 (p<0.001)

    • Plusoptix - SE (Specifically for Hyperopia, n=672): AUC = 0.856 (p<0.001) - Demonstrates much better performance when targeted at a specific condition.

    • Spot - SE (Specifically for Hyperopia, n=707): AUC = 0.886 (p<0.001) - Also shows strong performance for hyperopia.

    • PVS (Specifically for Strabismus, n=699): AUC = 0.654 (p<0.01)

The Role of Cut-off Values

  • Impact on Metrics: The chosen "cut-off" values or thresholds for a screening test directly influence its resulting sensitivity and specificity.

  • Arbitrary Guidelines: Cut-off values are often arbitrary guidelines that must be carefully decided.

  • Evidence-Based Decision-Making: Ideally, these cut-offs should be based on robust evidence to optimize the test's utility.

  • Example: Determining a grade cut-off for students to qualify as Teaching Assistants (TAs) is an analogous situation where an arbitrary but evidence-informed decision is made.

Conclusion on Screening Efficacy

  • Effectiveness: Kindergarten vision screening is an effective method for detecting vision problems in young children. While specificity may sometimes be lower, the high sensitivity in catching issues is crucial.

Contextual Considerations: Sensitivity vs. Specificity

The relative importance of sensitivity and specificity can vary significantly based on the healthcare and economic context.

  • Universal Medical Coverage (e.g., OHIP in Ontario):

    • In systems with universal coverage, false positives are generally less concerning than false negatives.

    • It is more desirable to over-refer (have some false alarms) to ensure that all actual problems are caught, as the cost of follow-up diagnosis is covered.

    • Therefore, sensitivity is often considered more important than specificity in such contexts.

  • US Context (Paid Medical Services):

    • In systems where patients or school boards bear the costs of medical services, specificity becomes more important.

    • School boards, for instance, would prefer fewer false positives to avoid the costs associated with unnecessary referrals, diagnoses, and treatments for children who do not actually have a vision problem.

Feasibility of Program Implementation

Beyond efficacy, the practical feasibility of scaling up a vision screening program is vital.

  • Research on Feasibility: Nishimura et al. (2020) published research in CMAJ on the feasibility of a school-based vision screening program in Ontario.

    • This study examined the practical challenges and successes of implementing such a program across various locations in Ontario.

  • Program Results (Nishimura et al., 2020):

    • Consent Models:

      • Using a passive-consent model (where parents must actively opt out if they do not want their children screened), a high rate of 89\% of children were screened.

      • In contrast, an active consent model (where parents must actively opt in) resulted in a lower screening rate of 62\%.

    • Referral Rates: Referral rates varied significantly between schools, ranging from 25-83\% for Junior Kindergarten (JK) children and 12-61\% for Senior Kindergarten (SK) children.

    • Problem Detection: Visual problems were detected in 10.7\% of the 4811 children screened.

    • First Eye Exam: For a remarkable 67\% of children identified with a visual problem through the screening, this was their very first eye examination, highlighting the program's role in reaching underserved populations.

  • Conclusion on Feasibility: Many children in Ontario have undiagnosed eye problems. In-school vision screening with follow-up eye examinations is an effective strategy for identifying at-risk children and ensuring they receive care before Grade 1, especially given the success of passive consent models.

Policy Impact and Evidence-Based Strategy

  • Policy Change: Building on the evidence from such research, in 2018 in-school vision screening was officially added to the public health standards for kindergarten children by the Ontario Ministry of Health and Long-Term Care.

  • Evidence-Based Strategy: This successful implementation serves as a prime example of an evidence-based school strategy, where scientific research directly informs and leads to significant public health policy changes.