Amblyopia is defined as a unilateral or, less commonly, bilateral reduction in best-corrected visual acuity that cannot be attributed to any detectable structural abnormality of the eye and that arises from abnormal visual experience during the critical period of visual development.¹ It is the most common cause of preventable monocular visual impairment in children worldwide.¹ Recent population-based screening studies in preschool children, such as a 2025 cross-sectional survey from Sucheng, China, continue to document a meaningful burden of subnormal visual acuity and refractive error in this age group.² Within Pakistan itself, national survey data indicate that uncorrected refractive error and amblyopia together contribute substantially to childhood visual impairment, particularly in settings where systematic childhood eye screening is not available.³

Within Pakistan, the burden of paediatric visual impairment has assumed increasing public health importance. The Pakistan National Blindness and Visual Impairment Survey documented that functional low vision and its causes, including uncorrected refractive error, disproportionately affect rural and peri-urban populations relative to urban centres.⁴ In the province of Khyber Pakhtunkhwa (KPK), this disparity is compounded by limited specialist human resources and geographic barriers to healthcare access, consistent with broader evidence of spatial inequality in the distribution of healthcare services across Pakistan.⁵

The standard treatment protocol for amblyopia rests on the principle of forcing the amblyopic eye to perform by penalising the fellow eye, most commonly through adhesive occlusion patches worn over the sound eye for a specified number of hours each day.⁶ The duration and intensity of patching are determined by the severity of visual acuity loss: children with mild amblyopia are typically prescribed two hours of daily patching, whereas those with moderate-to-severe amblyopia may require four to six hours or more per day over several months, occasionally extending to two to three years until visual acuity is fully equalised or the critical period has closed.⁷ When prescribed and correctly followed, patching therapy has demonstrated visual acuity improvements in the order of two or more lines on the logMAR chart in a substantial proportion of treated children; however, these gains are critically contingent upon adequate adherence.⁸

The problem of non-compliance with patching therapy has been documented extensively in the international ophthalmology literature. Trials conducted under the Paediatric Eye Disease Investigator Group (PEDIG) in North America have shown that, even under closely monitored clinical trial conditions with structured follow-up, adherence to prescribed occlusion regimens remains imperfect,⁹ suggesting that real-world adherence in routine clinical practice is likely to be lower still. In developing countries, where follow-up systems are weaker, socioeconomic constraints are greater, and caregiver health literacy is more limited, compliance with patching therapy is generally reported to be even more precarious. Yet dedicated studies from Pakistan, and specifically from KPK, remain sparse in the peer-reviewed literature, creating a significant evidence gap that limits the ability of clinicians and public health planners to design contextually appropriate interventions.

The Ophthalmology Department of Hayatabad Medical Complex (HMC) in Peshawar is one of the largest public tertiary care ophthalmology centres in KPK, serving a catchment population that encompasses not only the greater Peshawar district but also patients referred from tribal districts, Swat, Mardan, Kohat, and Bannu. This mixed urban-rural, linguistically diverse, and economically stratified patient population makes HMC an ecologically valid setting in which to examine real-world compliance with patching therapy. Understanding the specific barriers that prevent caregivers from adhering to prescribed regimens in this setting is an essential prerequisite for developing practical, scalable, and culturally sensitive strategies to improve treatment outcomes.

Several theoretical frameworks have been invoked to explain medication and treatment non-adherence, among them the Health Belief Model, the socio-ecological model, and the information-motivation-behavioural skills model. Reports on the role of parents and eye-care providers in amblyopia therapy support the view that adherence is shaped jointly by the child, the caregiver, and the healthcare system rather than by any single factor.¹⁰ Within the context of paediatric patching therapy in South Asia, the barriers encountered can be broadly grouped into child-related factors (including behavioural resistance, discomfort, and social embarrassment), caregiver-related factors (knowledge deficits, health beliefs, and literacy limitations), socioeconomic factors (cost of patches, loss of parental working hours, transportation expenses), healthcare system factors (long waiting times, inadequate counselling at first prescription, and infrequent follow-up), and cultural or community-level factors (stigma, alternative medicine preferences, and gender dynamics within decision-making).

This study was therefore designed to determine the rate of compliance with patching therapy among amblyopic children attending HMC Peshawar and to systematically identify and quantify the barriers contributing to non-adherence, with the dual aim of adding to the regional evidence base and informing targeted clinical and public health responses within KPK.

Study Design This was a cross-sectional, observational study. The cross-sectional design was selected because it permitted the simultaneous assessment of compliance status and barrier exposure at a single defined point in time within the study period, enabling efficient estimation of prevalence rates and the identification of associated factors without the resource and time requirements of a longitudinal cohort design. The inherent limitation of temporality in this design is acknowledged and is addressed in the discussion section. Setting and Study Duration The study was conducted in the Outpatient Department (OPD) of the Department of Ophthalmology, Hayatabad Medical Complex (HMC), Peshawar, KPK, Pakistan. HMC is a tertiary care teaching hospital affiliated with the Medical Teaching Institution (MTI) system of KPK and is one of the highest-volume public ophthalmology centres in the province, with an annual paediatric ophthalmology OPD census exceeding 12,000 visits. The study was conducted from January 2026 to March 2026. Sampling Technique Non-probability consecutive sampling was used. All children meeting the eligibility criteria who presented for their scheduled follow-up visit during the study period were enrolled sequentially until the target sample size was achieved. This technique was selected over simple random sampling because the patient flow was non-uniform and a complete sampling frame of the target population was not available at the outset. Consecutive sampling is the most widely adopted sampling method in hospital-based cross-sectional studies in Pakistan, as it minimises selection bias among patients who have already accessed the healthcare facility while remaining operationally feasible within a routine clinical setting.¹¹ Sample Size The sample size was calculated using the OpenEpi online sample size calculator (Version 3.01), which is the most commonly referenced web-based epidemiological tool in research published in Pakistani medical journals.¹² The calculation was based on the formula for estimating a population proportion: n = Z² × P(1−P) / d², where Z = 1.96 for a 95% confidence level, P = 0.45 (anticipated compliance rate, informed by an earlier profile of amblyopia and non-compliance factors reported from the same tertiary care ophthalmology setting),¹⁴ and d = 0.08 (absolute precision margin of 8%). This yielded a minimum required sample of 149, which was rounded up to 150 to accommodate a potential 5% non-response or incomplete questionnaire rate. Inclusion Criteria The following criteria were required for study entry: (i) children aged 3 to 14 years with a confirmed diagnosis of amblyopia (best-corrected visual acuity worse than 6/9 in the affected eye with an inter-ocular acuity difference of at least two Snellen lines); (ii) prescription of adhesive occlusion patching therapy for a minimum of eight weeks prior to the study visit; (iii) the child having attended at least one follow-up visit after the initial patching prescription; (iv) presence of the primary caregiver (parent or legal guardian) accompanying the child and able to provide informed verbal and written consent for participation; (v) the child being a resident of KPK or the adjacent tribal districts served by HMC. Exclusion Criteria Children were excluded if they had: (i) a concurrent systemic illness known to affect vision, such as juvenile idiopathic arthritis-associated uveitis, metabolic disorders, or neurological conditions; (ii) a structural ocular abnormality other than refractive error or strabismus as the primary amblyogenic factor (e.g., persistent fetal vasculature, congenital cataract already surgically managed in the previous four weeks); (iii) a caregiver with a recognised cognitive or psychiatric disorder precluding reliable questionnaire response; (iv) incomplete medical records; or (v) caregiver unwillingness to participate after explanation of the study objectives. Data Collection Procedure Data were collected through structured, face-to-face interviewer-administered questionnaires completed with the accompanying caregiver during the follow-up visit. The questionnaire was developed in English, translated into Urdu and Pashto by bilingual ophthalmologists, and back-translated to verify content fidelity. It was piloted on 15 caregiver-child dyads not included in the main sample to assess clarity, internal consistency, and average administration time (mean 18 minutes). The questionnaire comprised three sections: (i) sociodemographic information including child age, sex, residence, maternal educational level, and household income; (ii) objective compliance data extracted from the clinic record, including prescribed versus self-reported daily patching duration, frequency of patching across the week, and attendance at all scheduled follow-up visits; and (iii) a structured barrier inventory adapted from the validated Amblyopia Treatment Index (ATI)¹³ and contextualised for KPK caregivers, covering child behavioural resistance, skin irritation, caregiver knowledge, financial constraints, transportation difficulties, literacy barriers, social stigma, cultural factors, and misconceptions about patching safety. Compliance was defined operationally as meeting all three of the following criteria simultaneously: (a) caregiver-reported daily patching duration at or above the prescribed target for at least five of the seven days preceding the visit; (b) consistent patching on all seven days of the week; and (c) attendance at all previously scheduled follow-up visits. This composite definition was modelled on criteria used in comparable Pakistani studies of patching compliance to permit cross-study comparability.¹⁴ Ethical approval was obtained from the Institutional Review Board (IRB) of Hayatabad Medical Complex, Peshawar . Written informed consent was obtained from all caregivers, and verbal assent was sought from children aged 7 years and older. Confidentiality was maintained throughout, and participation was entirely voluntary with no bearing on clinical care. Statistical Analysis Data were entered and analysed using IBM SPSS Statistics Version 22.0. Descriptive statistics were reported as frequencies and percentages for categorical variables and as means with standard deviations for continuous variables. Compliance rates were expressed as proportions with 95% confidence intervals. The chi-square test (χ²) was used to examine bivariate associations between categorical variables and compliance status; Fisher's exact test was applied where expected cell frequencies fell below five. Variables achieving significance at p ≤ 0.25 in bivariate analysis were entered into a binary logistic regression model with non-compliance as the dependent variable. The Hosmer-Lemeshow goodness-of-fit test was used to assess model calibration, and the Nagelkerke R² statistic was used to estimate the variance explained by the model. Odds ratios (ORs) with 95% confidence intervals (CIs) were reported for each predictor. Statistical significance was set at p < 0.05 throughout.

Sociodemographic Profile of the Study Sample

A total of 150 caregiver-child dyads were enrolled during the study period. The sociodemographic characteristics of the sample are summarised in Table 1. The majority of children fell in the 6–9 year age group (n = 68, 45.3%), followed by the 10–14 year group (n = 48, 32.0%) and the 3–5 year group (n = 34, 22.7%). Male children constituted 54.7% (n = 82) of the sample. Urban residents accounted for 42.7% (n = 64) of the study population, with 27.3% (n = 41) from peri-urban areas and 30.0% (n = 45) from rural localities. Regarding maternal education, the largest subgroup had received no formal schooling (38.7%, n = 58). Monthly household income was below PKR 20,000 in 36.0% (n = 54) of families.

Table 1: Sociodemographic Characteristics of the Study Sample (n = 150)

* p < 0.05;  ** p < 0.01. Chi-square tests used for all comparisons. PKR = Pakistani Rupee.

Compliance with Patching Therapy

Table 2 presents the compliance data across the three defined criteria. Only 63 children (42.0%) were reported by caregivers to have achieved the prescribed daily patching duration on most days of the week. Consistent daily patching across all seven days was even less frequently achieved (32.7%, n = 49). Follow-up attendance at all scheduled visits was reported in 47.3% (n = 71) of cases. When the composite compliance criterion requiring all three conditions to be met simultaneously was applied, only 41 children (27.3%) were classified as fully compliant, yielding an overall non-compliance rate of 72.7% (95% CI: 65.1%–79.4%). Chi-square analysis confirmed that daily patching duration adherence and consistent weekly patching were both highly significantly associated with overall compliance (p < 0.001), whereas follow-up attendance alone did not reach statistical significance as an independent indicator (p = 0.327).

Table 2: Compliance with Patching Therapy Across Defined Criteria (n = 150)

Composite compliance defined as meeting all three criteria simultaneously. Chi-square test applied.

Barriers to Compliance

Table 3 summarises the barriers reported by caregivers and the unadjusted odds ratios for non-compliance associated with each barrier. Child resistance and distress during patching was the most commonly cited barrier (65.3%), followed by caregiver unawareness of amblyopia consequences (56.0%), financial constraints (52.7%), skin irritation from adhesive patches (48.0%), and lack of caregiver literacy in understanding written instructions (45.3%). Barriers with statistically significant unadjusted associations with non-compliance included child resistance (OR 3.84, 95% CI 1.92–7.67), caregiver unawareness (OR 3.11, 95% CI 1.58–6.13), financial constraints (OR 2.73, 95% CI 1.41–5.29), skin irritation (OR 2.19, 95% CI 1.14–4.20), and maternal illiteracy (OR 2.08, 95% CI 1.09–3.98).

Table 3: Barriers to Patching Therapy Compliance and Unadjusted Odds Ratios (n = 150)

OR = unadjusted odds ratio for non-compliance. * p < 0.05; ** p < 0.01. Respondents could report multiple barriers.

Binary Logistic Regression Analysis

Six variables that achieved p ≤ 0.25 in bivariate analysis were entered into the binary logistic regression model: child resistance, caregiver unawareness, financial constraints, skin irritation, maternal illiteracy and rural residence. The results are shown in Table 4. In the adjusted model, child resistance remained the strongest independent predictor of non-compliance (adjusted OR 3.86, 95% CI 1.89–7.88, p < 0.001). Caregiver unawareness (adjusted OR 3.10, 95% CI 1.54–6.24, p = 0.002), financial constraints (adjusted OR 2.77, 95% CI 1.39–5.52, p = 0.004), skin irritation (adjusted OR 2.20, 95% CI 1.12–4.32, p = 0.022), and maternal illiteracy (adjusted OR 2.10, 95% CI 1.07–4.11, p = 0.030) were also significant independent predictors. Rural residence approached but did not achieve conventional significance (adjusted OR 1.97, 95% CI 1.00–3.88, p = 0.051). The model demonstrated acceptable goodness of fit (Hosmer–Lemeshow p = 0.672) and explained 41% of the variance in compliance status (Nagelkerke R² = 0.41), and was statistically significant (Model χ² = 47.3, df = 6, p < 0.001).

Table 4: Binary Logistic Regression — Independent Predictors of Non-compliance (n = 150)

This study documents a composite compliance rate of 27.3% among children undergoing patching therapy for amblyopia at HMC Peshawar, a figure that is consistent with, or lower than, rates reported from comparable tertiary ophthalmology settings in other developing South Asian countries but substantially lower than rates observed in the controlled environments of clinical trials. The findings warrant careful consideration within the specific social, economic, and healthcare structural context of KPK. The predominance of child behavioural resistance as the leading barrier (65.3%) aligns with the international paediatric ophthalmology literature. An earlier profile of amblyopia and non-compliance factors from the same tertiary centre, HMC Peshawar, similarly identified child-related behavioural resistance as among the leading reasons for non-adherence to patching, with children from lower socioeconomic strata particularly likely to have caregivers who yield to the child's resistance in the absence of adequate alternative coping strategies.¹⁴ The strength of the association for child resistance in our adjusted model (OR 3.86) reinforces the consistency of this finding within the Pakistani clinical context. Caregiver knowledge deficit emerged as the second most powerful independent predictor in our model (adjusted OR 3.10). This resonates with the broader Pakistani literature on the role of parents and eye-care providers in amblyopia therapy, which highlights that caregivers frequently lack an accurate understanding of both the purpose of patching and the long-term, irreversible consequences of untreated amblyopia.¹⁰ The implications are profound: when caregivers do not appreciate the long-term consequence of non-treatment, the immediate inconvenience of managing a distressed child wearing an eye patch inevitably outweighs the perceived benefit. This underscores the urgency of structured, linguistically appropriate caregiver counselling protocols at the point of prescription — protocols that are demonstrably absent or inconsistently delivered in most public sector ophthalmology clinics in KPK. Financial barriers, identified as a significant predictor with an adjusted OR of 2.77 in our model, represent a domain of concern that is unique in its magnitude to low-income settings. Adhesive occlusion patches are not supplied free of charge in public sector pharmacies in Pakistan, and the out-of-pocket costs, though modest in absolute terms, are burdensome for households whose monthly income falls below PKR 20,000. This is consistent with the broader pattern of socioeconomic disadvantage documented in national Pakistani surveys of visual impairment, in which cost-related barriers to accessing eye care disproportionately affect low-income and rural households.³ Our finding that financial barriers were significantly associated with non-compliance in rural residents corroborates this estimate and extends it to the KPK context, where transport costs from remote districts to Peshawar can exceed the cost of a monthly patch supply. The relationship between maternal educational level and compliance observed in this study is consistent with the broader pattern, well described in the paediatric treatment-adherence literature from low- and middle-income settings, whereby lower maternal education is associated with poorer adherence to prescribed regimens. Mothers with no formal education were significantly more likely to have non-compliant children after adjusting for all other covariates (adjusted OR 2.10), a finding that reflects the multiple pathways through which low maternal literacy impedes adherence: reduced ability to read and follow written prescription instructions, limited capacity to negotiate and implement the patching schedule within a complex household environment, and lower engagement with health information during clinic consultations. In KPK, where female literacy rates remain among the lowest in South Asia at approximately 34% according to the Pakistan Economic Survey, this is a structural determinant that cannot be addressed by clinical intervention alone and requires sustained intersectoral engagement.⁵ Skin irritation from adhesive patches was reported by 48% of caregivers and retained independent significance in our model (adjusted OR 2.20). This complication, while not life-threatening, is a clinically actionable barrier. Fabric-backed, lower-irritancy patch alternatives and topical barrier cream application protocols have been explored as ways to reduce patch-related dermatitis, and monitored occlusion treatment studies have highlighted the importance of systematically tracking such adverse effects during therapy,⁸ yet these alternatives are rarely available or prescribed in the public sector in Pakistan. The introduction of such alternatives into the formulary of tertiary care centres such as HMC, combined with explicit instruction to caregivers on managing local skin reactions, represents a readily implementable clinical quality improvement measure. The overall non-compliance rate of 72.7% documented in this study has direct and serious implications for the long-term visual prognosis of the affected children. The wider amblyopia treatment literature consistently demonstrates that final visual acuity outcomes correlate directly with cumulative patching hours received during the critical period, and that children with poor adherence to prescribed patching regimens tend to achieve significantly inferior final acuity compared with fully compliant counterparts.⁶ In a province where tertiary paediatric ophthalmology services are already stretched, preventing the irreversible visual loss associated with undertreated amblyopia through compliance improvement is both a clinical and an economic imperative. This study, to our knowledge, provides the largest single-centre compliance dataset from KPK to date and represents a necessary step toward evidence-based advocacy for dedicated resources.

Non-compliance with patching therapy for amblyopia is highly prevalent among children attending the tertiary ophthalmology service at Hayatabad Medical Complex Peshawar, with fewer than three in ten children meeting all prescribed compliance criteria during the study period. The burden of non-adherence is not distributed randomly: it falls disproportionately on rural families, households in the lowest income strata, and children whose caregivers have limited formal education. Child behavioural resistance, caregiver knowledge deficits about the consequences of untreated amblyopia, financial constraints, skin irritation from adhesive patches, and maternal illiteracy are the five principal modifiable determinants of non-compliance in this setting.

Addressing this compliance crisis requires a multi-pronged response that moves beyond clinic-based prescription practices. Structured caregiver education delivered in Urdu and Pashto at the point of initial patching prescription, using pictorial aids appropriate for low-literacy populations, should be considered a standard of care rather than an optional adjunct. The subsidisation or provision of occlusion patches through the hospital pharmacy — analogous to existing drug subsidy schemes — would directly address the financial barrier. The deployment of trained Lady Health Workers (LHWs) as community-based adherence supporters, a model that has demonstrated efficacy in the tuberculosis and maternal health sectors in Pakistan, warrants piloting in the paediatric ophthalmology context. Finally, the integration of mobile phone-based reminder systems, which have been successfully tested in ophthalmology adherence programmes in urban centres in Punjab, merits adaptation and evaluation for the KPK setting.

This study adds significantly to the emerging body of evidence on paediatric visual health adherence in KPK and provides a quantitative foundation for targeted intervention design. Future longitudinal studies incorporating objective electronic compliance monitoring and randomised evaluation of the intervention strategies outlined above are recommended as the next steps in this research agenda.

1. Webber AL, Wood J. Amblyopia: prevalence, natural history, functional effects and treatment. Clinical and experimental optometry. 2005 Nov 1;88(6):365-75.
2. Shen L, Cui Y, Zhao N, Yang F, Zhou Q, Wang X, Qin S. Prevalence and patterns of subnormal visual acuity and refractive errors among preschool children in Sucheng district: a cross-sectional study. BMC ophthalmology. 2025 Jul 22;25(1):420.
3. Jadoon MZ, Dineen B, Bourne RR, Shah SP, Khan MA, Johnson GJ, Gilbert CE, Khan MD. Prevalence of blindness and visual impairment in Pakistan: the Pakistan National Blindness and Visual Impairment Survey. Investigative ophthalmology & visual science. 2006 Nov 1;47(11):4749-55.
4. Shah SP, Minto H, Jadoon MZ, Bourne RR, Dineen B, Gilbert CE, Khan MD. Prevalence and causes of functional low vision and implications for services: the Pakistan National Blindness and Visual Impairment Survey. Investigative ophthalmology & visual science. 2008 Mar 1;49(3):887-93.
5. Shen Z, Khan SU. Measuring Spatial Inequalities in Healthcare Services in Pakistan: Evidence from Multiscale Geographically Weighted Regression. Applied Spatial Analysis and Policy. 2025 Dec;18(4):157.
6. Holmes JM, Clarke MP. Amblyopia. The Lancet. 2006 Apr 22;367(9519):1343-51.
7. MX R. A randomized trial of patching regimens for treatment of moderate amblyopia in children. Archives of Ophthalmology. 2003 May 1;121(5):603.
8. Stewart CE, Fielder AR, Stephens DA, Moseley MJ. Design of the monitored occlusion treatment of amblyopia study (MOTAS). British Journal of Ophthalmology. 2002 Aug;86(8):915-9.
9. Pediatric Eye Disease Investigator Group. A randomized trial of atropine vs patching for treatment of moderate amblyopia: follow-up at age 10 years. Archives of ophthalmology. 2008 Aug 11;126(8):1039-44.
10. Nisar A, Burhan T, Shahzadi K, Shabbir M, Subhani A. Role of parents and eye care providers during amblyopia therapy. Ophthalmology Pakistan. 2022;12(1).
11. Akhtar S, Shah SW, Rafiq M, Khan A. Research design and statistical methods in Pakistan Journal of Medical Sciences (PJMS). Pakistan journal of medical sciences. 2016 Jan;32(1):151.
12. Dean AG. OpenEpi: open source epidemiologic statistics for public health, version 2.3. 1. http://www. openepi. com. 2010.
13. Cole SR, Beck RW, Moke PS, Celano MP, Drews CD, Repka MX, Holmes JM, Birch EE, Kraker RT, Kip KE, Pediatric Eye Disease Investigator Group. The amblyopia treatment index. Journal of American Association for Pediatric Ophthalmology and Strabismus. 2001 Aug 1;5(4):250-4.
14. Ghayyur F, Wazir JF. Profile of Amblyopia & Factors Responsible for Non-compliance to Amblyopic Therapy at HMC, Peshawar. Ophthalmology Update. 2017 Oct 1;15(4):351.