Purpose
To measure the difference in Convergence Insufficiency Symptom Survey scores for reading vs favorite near visual activities.
Design
Comparative validity analysis of diagnostic tools.
Methods
At a single clinical private practice, 100 children aged 9–18 with normal binocular vision were recruited to receive either the original survey emphasizing reading or a modified survey replacing “reading” with their favorite near activity. Average survey scores and subscores for questions emphasizing fatigue, discomfort, impaired vision, and cognitive performance were compared using t tests, while responses to individual questions were compared using Mann-Whitney U tests.
Results
The average reading survey score was significantly greater than the favorite near activity survey score (14.1 ± 11.5 vs 6.7 ± 5.8, P = .0001). The largest difference resulted from questions emphasizing cognitive performance (subscore 5.8 ± 4.3 vs 2.0 ± 2.1, P = .0000002), although significant differences were also found for fatigue (5.4 ± 3.8 vs 3.0 ± 2.7, P = .0003), discomfort (3.9 ± 4.6 vs 1.8 ± 2.2, P = .004), and impaired vision (3.2 ± 3.9 vs 1.8 ± 2.2, P = .02). Significant differences were found for 7 survey questions, with higher symptom scores for the reading survey in every case. Using survey scores ≥16 to diagnose convergence insufficiency, significantly more children taking the reading survey would have been diagnosed with convergence insufficiency than children taking the favorite near activity survey (19 of 50 [38%] vs 5 of 50 [10%], P = .001).
Conclusions
By emphasizing reading, the Convergence Insufficiency Symptom Survey score significantly overestimates near visual symptoms in children with normal binocular vision compared with symptoms caused by preferred near activities that require similar amplitudes of accommodation and convergence.
Vision is the primary modality that children use to acquire new academic information. The extent to which vision problems contribute to learning disabilities, however, is controversial. The ophthalmic literature strongly implicates a minor role for vision problems in learning disabilities, particularly dyslexia, a language-based disorder involving neuroprocessing problems unrelated to vision. The optometric literature, on the other hand, strongly advocates a primary role for vision in impaired learning, dividing vision problems into “visual efficiency” disorders like poor convergence and accommodation and “visual information processing” disorders of higher cortical functions that both contribute to ineffective learning. The conflicting literature does reach a consensus in 1 area: abnormal convergence and/or poor accommodation interfere with near vision and thus negatively impact visual information processing. In particular, there appears to be a link between attention deficit hyperactivity disorder (ADHD) and convergence insufficiency (CI), although that association is controversial and may not be strong or conclusive. Despite the controversies in the literature, it is reasonable to conclude that uncomfortable near vision may be an impediment to learning.
The prevalence of CI is estimated to be 4%–13% in fifth- and sixth-graders. The Convergence Insufficiency Symptom Survey score was created to provide a quantitative measure of the symptoms caused by abnormally poor convergence ability. This score has been used in randomized controlled trials to measure the improvement in symptoms after various treatments for CI. Although it was designed as a research tool, not a diagnostic tool, its usage has recently been expanded to help diagnose CI by establishing a Convergence Insufficiency Symptom Survey score ≥16 as one of the criteria used to define CI in children.
A major problem with the Convergence Insufficiency Symptom Survey, however, is its emphasis on reading compared with other near activities, an emphasis that may cause an overestimation of CI symptoms in children with poor reading skills. This deficiency was recognized by the creators of the survey, leading to a modification of most questions to include the revised wording “reading or doing close work.” Reading is the only near vision task specified in all questions, however, and 3 of the 15 questions focus exclusively on reading. Other factors unrelated to poor near vision might contribute to a high symptom score, including the overlap of many Convergence Insufficiency Symptom Survey symptoms with ADHD. In children without the diagnosis of CI, fatigue and deficiencies of higher cognitive functions like concentration, comprehension, memory, and reading skills all were associated with higher symptom scores. A modification has been proposed to the original Convergence Insufficiency Symptom Survey to ask subjects to determine if their symptoms were caused by their eyes in an attempt to control for the influence of nonocular factors on the survey score.
Since most near visual activities require similar amplitudes of convergence and accommodation, the Convergence Insufficiency Symptom Survey score may be influenced both by the child’s motivation to perform the near activity and by the child’s internally perceived success at that activity, factors unrelated to near visual ability. To control for those factors, this study measured the difference between the symptom scores for the original reading-based Convergence Insufficiency Symptom Survey (reading survey) vs the symptom score for a modified favorite near visual activities Convergence Insufficiency Symptom Survey (favorite survey) in school-age children. This study also directly estimated the effects of potential confounding factors like fatigue and cognitive ability by grouping related questions to create subscores for those and other factors to determine their relative contributions to the overall Convergence Insufficiency Symptom Survey score.
Methods
This study was performed at a single clinical private practice as a prospective, randomized survey of children 9–18 years of age with normal binocular vision. Parents or guardians gave informed consent and all participants gave assent to a protocol conforming to the Declaration of Helsinki and prospectively approved by the Institutional Review Board at Providence Little Company of Mary Medical Center, Torrance, California, USA. The data collection was compliant with the Health Insurance Portability and Accountability Act.
Children were recruited if, on ophthalmology examination, they had corrected distance acuity of 20/25 or better OU, near acuity J1+ font at 13 inches OU, no detectable phorias or tropias at distance and near, and a near point of convergence 6 cm or closer. Exclusion criteria included history of ADHD or other self-reported learning disabilities, strabismus, ocular surgery, or developmental delay.
Copies of each survey were printed and then randomized using a random number generator, odd for the reading surveys and even for the favorite survey, to create a single large stack of intermingled surveys. Then, as each child was recruited, the survey sheet on top was used to record the responses for that child. The surveyor gave instructions and read all questions out loud to each participant, then recorded the results on the printed survey datasheet. The original Convergence Insufficiency Symptom Survey was modified to include a new introductory question ( Table 1 ), “What is your favorite activity up close that uses your eyes?” The answer to this question did not contribute to the symptom score. Instead, it was simply noted for the children randomized to receive the reading survey and it was used to modify the favorite survey for children randomized to be surveyed on their favorite near visual activity. For example, the most frequently chosen favorite near visual activity in both groups was use of a smartphone ( Table 2 ); in the favorite survey, “reading or doing close work” was directly replaced with “using your smartphone” for questions 1–5, 7, and 10–13 ( Table 1 ). Questions 6, 8, 9, and 14, however, required additional modification of the wording to properly capture the intent of the question. For example, question 9 of the original Convergence Insufficiency Symptom Survey asks, “Do you feel like you read slowly?” In the favorite survey, question 9 was changed to “Do you feel like you have a hard time completing/finishing your favorite near activity?” While reading the question, the surveyor modified the question further to conform to each child’s favorite near activity (eg, “Do you feel like you have a hard time completing activities on your smartphone?”). Because these questions required more substantial wording changes from the original Convergence Insufficiency Symptom Survey, the scores generated were analyzed separately in the results to ensure that the modified wording did not affect the study outcomes. Finally, no clear alternative wording for nonreading activities could capture the intent of question 15 in the original Convergence Insufficiency Symptom Survey, “Do you have to re-read the same line of words when reading?” This question was eliminated from both the reading survey and favorite survey in this study. In the original Convergence Insufficiency Symptom Survey study, question 15 only contributed an average of 0.86 points to the Convergence Insufficiency Symptom Survey score for children with normal binocular vision, so excluding this question should not substantially affect the symptom scores for either group.
| Introductory Question: What is your favorite activity up close that uses your eyes? | |
|---|---|
| Reading Questions | Favorite Questions |
| 1. Do your eyes feel tired when reading or doing close work ? | 1. Do your eyes feel tired when “doing your favorite near activity” ? |
| 2. Do your eyes feel uncomfortable when reading or doing close work ? | 2. Do your eyes feel uncomfortable when “doing your favorite near activity” ? |
| 3. Do you have headaches when reading or doing close work ? | 3. Do you have headaches when “doing your favorite near activity” ? |
| 4. Do you feel sleepy when reading or doing close work ? | 4. Do you feel sleepy when “doing your favorite near activity” ? |
| 5. Do you lose concentration when reading or doing close work ? | 5. Do you lose concentration when “doing your favorite near activity” ? |
| 6. Do you have trouble remembering what you have read? | 6. Do you have trouble remembering what you have done after finishing “your favorite near activity” ? |
| 7. Do you have double vision when reading or doing close work ? | 7. Do you have double vision when “doing your favorite near activity” ? |
| 8. Do you see the words move, jump, swim, or appear to float on the page when reading or doing close work ? | 8. Do you see near images move, jump, swim, or appear to float in front of each other when “doing your favorite near activity” ? |
| 9. Do you feel like you read slowly? | 9. Do you feel like you have a hard time “completing/finishing your favorite near activity” ? |
| 10. Do your eyes ever hurt when reading or doing close work ? | 10. Do your eyes ever hurt when “doing your favorite near activity” ? |
| 11. Do your eyes ever feel sore when reading or doing close work ? | 11. Do your eyes ever feel sore when “doing your favorite near activity” ? |
| 12. Do you feel a “pulling” feeling around your eyes when reading or doing close work ? | 12. Do you feel a “pulling” feeling around your eyes when “doing your favorite near activity” ? |
| 13. Do you notice the words blurring or coming in and out of focus when reading or doing close work ? | 13. Do you notice your vision blurring or coming in and out of focus when “doing your favorite near activity” ? |
| 14. Do you lose your place while reading or doing close work ? | 14. Do you lose track of what is happening while “doing your favorite near activity” ? |
| Reading Survey | Favorite Survey | P Value | |
|---|---|---|---|
| Sex (male/female) | 22/28 | 27/23 | .32 |
| Average age (y) | 12.7 ± 2.5 | 12.6 ± 2.5 | .81 |
| Favorite activity | χ 2 | df 7, 5.144 | .64 |
| Smartphone | 16 | 12 | |
| Reading | 13 | 10 | |
| Tablet | 8 | 6 | |
| Computer | 2 | 4 | |
| Video gaming | 3 | 7 | |
| Videos | 3 | 2 | |
| Drawing | 2 | 5 | |
| Other a | 3 | 4 |
To create subscores, questions were grouped based on the following possible confounding factors: fatigue (questions 1, 4–6), discomfort (questions 2, 3, 10–12), impaired vision (questions 2, 7, 8, 12, 13), and cognitive performance (questions 5, 6, 9, 14). Because some questions were associated with multiple subscores, the summation of the average individual subscores should be greater than the overall average symptom score for each group.
A power analysis determined that recruiting 50 subjects per group gave the study a 90% probability of detecting a 4-point difference in average symptom scores between groups at P < .05. For statistical analysis, means, standard deviations, and t tests were calculated for summed scores; Mann-Whitney U tests were used to compare responses to individual survey questions; and χ 2 tests were used to compare percentage differences between groups using Microsoft Excel for Macintosh 2011 (Microsoft Corporation, Redmond, Washington, USA) enhanced with the Real Statistics Resource Pack (Charles Zaiontz, Trento, Italy).
Results
Every child who passed the exclusion criteria with normal binocular vision on examination, regardless of presenting complaint, was recruited for the study and every recruited child and parent agreed to participate with the survey. Both groups of 50 children had a similar distribution of sexes, ages, and favorite near visual activities ( Table 2 ). The average symptom score was significantly greater for the reading survey than the favorite survey ( Figure , 14.1 ± 11.5 vs 6.7 ± 5.8, P = .0001). Similarly, the reading survey score was significantly greater than the favorite survey score after excluding questions 6, 8, 9, and 14 (9.6 ± 8.6 vs 5.4 ± 4.8, P = .003), demonstrating that modification of those questions did not significantly alter the relative difference in scores between the 2 surveys. Cronbach’s alpha was calculated to check for the internal consistency of responses within the reading survey, within the favorite survey, and again within both surveys after excluding questions 6, 8, 9, and 14; the values ranged from 0.72 to 0.89, evidence of good internal consistency among the questions within all of the surveys with or without modifications. Finally, 23 children chose reading as their favorite near activity, 13 for the reading survey and 10 for the favorite survey ( Table 2 ). These children had nearly identical average symptom scores (11.4 ± 14.1 vs 10.7 ± 7.4, P = .89), further evidence that modifying the wording of the questions did not significantly affect the overall symptom scores.
To allow a direct comparison of symptom scores for reading vs other favorite near visual activities, the children who chose reading as their favorite activity were removed from both groups and the average scores were recalculated. For the remaining 77 children, the average reading survey score was almost triple the average favorite survey score (15.0 ± 10.5 vs 5.8 ± 5.0, P = .000004). Even children who chose reading as their favorite near visual activity reported symptom scores significantly higher than children who chose other favorite activities (11.1 ± 11.5 vs 5.8 ± 5.0, P = .01). Using a cutoff Convergence Insufficiency Symptom Survey score ≥16, the reading survey would have falsely diagnosed CI in 19 children (38%), while the favorite survey would have falsely diagnosed CI in 5 children (10%) (χ 2 df 1, 10.75, P = .001). Three subjects in each group had symptom scores >16 even though they selected reading as their favorite near visual activity, demonstrating that a high symptom score did not preclude children’s enjoyment of reading.
It is possible that the use of nonreading near visual activities might underestimate near visual symptoms compared with reading. Using the computer, playing video games, and watching videos all might occur at a more remote distance from the eyes than reading material is typically held. These activities also might involve larger visual targets with less detail than letters and words. In this study, the near visual activity chosen that was most directly comparable to reading was the use of handheld electronics like smartphones and tablets, typically viewed at a similar distance from the eyes as reading material and involving a similar level of visual detail. The 42 children who listed handheld electronics as their favorite near visual activity reported nearly identical scores for both the reading survey and favorite survey as the overall groups (14.3 ± 10.3 vs 6.3 ± 4.3, P = .003). Thus, the viewing distance of the near visual activity did not appear to affect the symptom scores.
The comparisons of individual questions are summarized in Table 3 . For 7 questions, the reading survey had significantly higher symptom scores than the favorite survey, while the remaining 7 questions generated similar symptom scores. For the reading survey, only question 5 (about losing concentration) had a mode >0. Six other questions had median responses of 1 or 2. For the favorite symptom survey, every question had a mode of 0 and only 2 questions had a median response of 1, question 1 (about feeling tired) and 13 (about blurred vision).
