"Outcome Evaluation of the Visual Preschool Screening in the Province of Trento North East Italy"

Amblyopia represents the main cause of visual impairment in the developmental age, with a prevalence in preschool children...


Introduction
Amblyopia represents the main cause of visual impairment in the developmental age. It is a condition of reduced mono or bilateral visual acuity that also occurs independently of an organic cause. It is due to inadequate visual stimulation during the plastic period of development of the visual system. Literature data indicate a prevalence of amblyopia in preschool age between 1.5 and 4.0%. Its most common causes are represented by strabismus and refractive errors [1]. In developing countries, amblyopia is the second most common cause of functional vision impairment in children [2]. Preschool visual screening allows for an early diagnosis of amblyopia and consequently an early treatment which, if implemented within 6 years, can allow a recovery, even total, of the visual function. This aspect is very important for the overall development of the child and his / her abilities / possibilities for learning and social integration [3,4].
Given the need to treat the basic conditions (any congenital cataracts, refractive errors, etc.), the classic treatment essentially rests on two approaches: the eye bandage that has better visual acuity (for a few hours a day depending on the severity of the amblyopia and / or the tolerance of the treatment), in order to work the amblyopic eye or, alternatively or in addition to the use of penalties such as filters on the lenses, less coercive treatment than bandages, but which does not preclude the possibility of binocular collaboration. The duration of treatment depends on the extent of amblyopia and compliance with therapy and should be modulated on the results of periodic checks to be carried out in the followup after the start of treatment [5][6][7][8][9]. This screening represents an evidence-based practice, as recently reconfirmed by the US Preventive Service Task Force, a body of the American Federal Agency for Research and Quality of Care.
The latest update, released in 2017, confirms the adequacy of screening between the ages of 3 and 5, while robust evidence of efficacy for screening in children under the age of three is not yet available [10]. The data on screening are also satisfactory in terms of cost-effectiveness [11]. Preschool visual screening, in the second year of kindergarten, has been active in the province of Trento (540,000 inhabitants in north-east Italy) since the second half of the 1980s and currently covers the entire province with long-established organizational and operational criteria. The organizational aspects, operating procedures and coverage of the screening in place in the province of Trento were the subject of previous work [12]. This study reports on the evaluation of the final outcome of orthotic screening in three consecutive cohorts of children undergoing screening and diagnosed with amblyopia at the age of 4, considering the quality of vision verified after 3 years.

Material and Methods
The subjects identified as amblyopic, born in the years 2008-2010 and subjected to pre-school visual screening at 4 years (second year of kindergarten), that is in the years 2012-2014, were taken into consideration. These subjects, registered in the screening archive, were followed retrospectively up to three years after the date of screening, then at the age of 6-7, to verify whether or not they had undergone a control eye examination, verifying the visual acuity values achieved for the right eye and the left eye respectively. The data relating to the treatment carried out were also recovered. The recovery of follow-up data was carried out than those of other charts [13,14]. The significance of the differences between the proportions in comparison was tested by the chisquared test or Fisher's exact test, as needed. The proportions are provided by 95% confidence intervals (95% CI).

Results
In the years 2012-2014, 13,638 children underwent preschool visual screening, with an average coverage (net of subjects already in care, of those absent from school and not recovered subsequently and of those who did not receive parental consent) equal to 96%. In the period under study, an average of two children per year were identified as having amblyopia within the first three years of life, therefore before the invitation to preschool screening.
Children classified as amblyopic were 208 at the end of the various stages of screening, for an average prevalence for the period of 1.5% (Table 1). Bilateral cases of amblyopia were 10 (4.8%) and severe forms equal 8 cases (3.8%). Males represent 51.9% of the series (108 subjects) and children of foreign citizenship 13.9% (29 subjects). The visual acuity level for the two eyes at screening, for the 208 cases diagnosed with amblyopia, is shown in ( Table 2). The difference in visual acuity between the two eyes is shown in (Table   3).   The subdivision according to the refractive defects identified is shown in Table 4, from which it emerges that the most frequent condition is astigmatism which affects 80% of cases.
In 203 cases (97.6%) corrective lenses were prescribed, at the end of the screening, to ascertain an amblyopia condition by the Of the latter cases, 4 were also related to subjects who moved out of the province after the completion of the screening and therefore not assessable for the effectiveness of the treatment. The actual proportion of those assessed remotely is therefore equal to 87.4%, of the subjects actually assessable. The proportion of subjects for whom the information was retrieved did not vary in a statistically significant way, in relation    The data relating to myopic people is also emphasized by the low number of cases: 10 evaluated at 6 years compared to the total of 13 identified at screening. The data relating to subjects with overt squint is the only statistically significant with respect to the mean (p <0.001). Finally, in only three cases the visual acuity in the amblyopic eye is equal to 4 tenths (> 0.30 LOGMar).

Discussion
A long-term outcome evaluation is the only reliable way to evaluate the effectiveness of a screening. This is also valid for preschool visual screening where the goal is to improve the levels of visual acuity to support the full development of the child and his or her learning and social integration possibilities. Our study represents the first outcome assessment of the preschool visual screening program in the province of Trento, more than thirty years after its launch. A shorter periodicity of evaluation would be recommended in order to understand in a short time whether the program works or not. The evaluation activity is however onerous in terms of time expenditure and requires the availability of an information system that allows the useful data to be available. The evaluation would also require, to be more appropriate, to have data on the child's educational impact and for this reason it should consider a time well beyond the 6-7 years threshold.
Anyway, SIO, in our experience, appears to be a useful information tool even if it could be supplemented by information that can be provided by the family pediatricians. These professionals follow children from birth through puberty so they are ideally placed to provide prospective data. Their involvement in this type of activity should in any case be regulated and appropriately encouraged. An alternative way could be to contact directly the family even though this approach may not provide all the data of interest [11,15,16].
Defining any preferential ways for the specialist assessment of screened subjects could in any case be advantageous, also for the purpose of retrieving information for outcome assessments [11].
Referring to the current available information flows we were able to recover data relating to a post-screening evaluation for 87.4% of cases (about 9 cases out of 10). An incomplete recovery of the follow-up data could be related, rather than to an actual lack of adherence to the therapeutic path, to the failure to register remote assessments in the SIO, or to access to private eye clinics, which implies an inability to access the evaluation of the treatment given that the services in the private sector are not registered in the SIO.
This may explain the fact that foreign children are more likely to be assessed remotely than Italians, as the former tend more frequently to access public facilities than the seconds. On the basis of the available data, we report no differences in access to the 6-7th year outcome assessment in relation to gender, citizenship and residence area. Even in the presence of incomplete evaluation data, a picture of homogeneity and equity of the treatment path emerges without the presence of apparent barriers, which have already been reported in the literature [17][18][19]. However, precise data relating to compliance with therapy are not available, an aspect of great importance for the purposes of the efficacy of the treatment and which is influenced by the level of awareness of parents on the importance of regularity of periodic checks, post screening [11,[20][21][22]. Compliance with treatment can be guaranteed on the basis of optimal collaboration between the family and the care team: orthoptic / eye services and family pediatrician.
A qualitative study, designed ad hoc, would have been indicated for the assessment of compliance [11]. From our study we can hypothesize that at least 16 children (9.0%) may have had problems accepting the occlusive treatment. We can hypothesize greater compliance problems in children with a foreign mother, considering that the effectiveness achieved in foreigners is lower than in Italians even if the difference is not statistically significant.
Problems with access to services and compliance with treatment in the foreign population have also been reported in the literature [23,24]. Finally, we found no differences in relation to the mother's level of education, an aspect indicated as a possible barrier to access to treatment in other studies [25]. Taking into account the information limits mentioned above, we can believe that the pre-school visual screening program of the province of Trento is effective. A clear increase in visual acuity emerges, both for individuals and for all the subjects evaluated.
The differences between the visual acuity values at 6-7 years and the screening values appear statistically significant, in line with what is reported in the literature [26][27][28][29][30][31][32], despite the difference in programs and procedures. There are no relevant associations between the results at 6-7 years and the degree of amblyopia and/or the refractive defect as identified at screening even if in astigmatics the improvement in visual performance seems better.
In conclusion, the remote results of the provincial preschool visual