Patients’ knowledge, attitudes, and practices toward dry eye disease management Authors: Chongyan Zhu, Yajie Zhou, Hongwei Zhou Published: 2025 ABSTRACT AbstractBackgroundThis study aimed to assess the knowledge, attitudes and practices (KAP) among patients with dry eye disease (DED) toward disease management.MethodsThis web-based cross-sectional study was conducted between February 2023 and August 2024 at our hospital. A self-designed questionnaire was developed to collect demographic information of DED patients, and assess their KAP toward disease management.ResultsA total of 500 valid questionnaires were collected. Among them, 58.60% (293) were male, 26.80% (134) aged 43–54 years old and 51.2% (256) with moderate severity of DED. The mean score for knowledge, attitudes and practices was 5.80 ± 2.39 (possible range: 0–11), 38.68 ± 5.26 (possible range: 11–55), and 29.60 ± 8.22 (possible range: 9–45), respectively. Pearson’s correlation analysis revealed positive correlation between knowledge and attitudes (r = 0.482, P < 0.001). Multivariate logistic analysis demonstrated that being divorced or widowed (OR = 0.382, 95% CI: 0.165–0.886, P = 0.025) and moderate dry eye severity (OR = 0.459, 95% CI: 0.269–0.781, P = 0.004) were independently associated with knowledge, possibly due to limited social support and lower health literacy among these subgroups. Knowledge (OR = 1.456, 95% CI: 1.308–1.621, P < 0.001), moderate dry eye severity (OR = 0.452, 95% CI: 0.254–0.804, P = 0.007) and occasionally experiencing eye discomfort in the past week (OR = 0.431, 95% CI: 0.189–0.982, P = 0.045) were independently associated with attitudes. Attitudes (OR = 1.082, 95% CI: 1.033–1.134, P = 0.001), being divorced or widowed (OR = 0.447, 95% CI: 0.211–0.944, P = 0.035), aqueous deficiency (OR = 2.103, 95% CI: 1.036–4.272, P = 0.040) and lipid abnormality (OR = 2.937, 95% CI: 1.616–5.338, P < 0.001) were independently associated with practices. Structural equation model demonstrated that knowledge positively affect attitudes (path coefficient = 2.950, P < 0.001) and attitudes positively affect practices (path coefficient = 0.595, P < 0.001).ConclusionDED patients demonstrated insufficient knowledge, passive attitudes, and poor practices toward DED. The implementation of educational initiatives and comprehensive training programs emerges as imperative in order to enhance KAP among DED patients.Clinical trial numberNot applicable.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12886-025-04520-y.Keywords: Knowledge, Attitudes, Practices, Dry eye disease, Cross-sectional study Background Dry eye disease (DED) is a prevalent ocular condition characterized by symptoms such as ocular discomfort, visual disturbances, and tear film instability [1]. The persistent discomfort and visual disturbances associated with DED significantly diminish patients’ quality of life, impairing their ability to engage in daily activities [2, 3]. Beyond physical health concerns, ocular discomfort (such as pain, dryness, or visual disturbance) and visual impairment can also lead to increased psychological distress [4]. In the context of societal shifts towards increased screen time, environmental factors, and aging populations, the prevalence of DED has witnessed a remarkable rise in recent years [5–7]. The global prevalence of DED is approximately 11.59%, the prevalence of DED in the Chinese population aged 5–89 years was 13.55%, affecting a total of 170.09 million individuals [8]. Moreover, among Chinese high school students, the prevalence of symptomatic DED was alarmingly high at 70.5% [9]. Therefore, the imperative for preventative measures, early diagnosis, and effective treatment strategies becomes evident. DED patients’ awareness of disease symptoms plays important roles in the disease management and the deficiency in DED related knowledge can impede the proactive treatment of it [10, 11]. Moreover, the attitudes the patients held for DED greatly augment their compliance to routine follow-up visits and enhances their confidence in medical treatment [12]. Furthermore, adherence to clinical diagnosis and therapeutic interventions, such as artificial tear supplements and anti-inflammatory medications enables the optimization of patients’ prognosis and alleviation of DED burden [13]. Currently, published researches of DED have mainly focused on healthcare personnel [14], and there is a dearth of investigations among DED patients. To comprehensively understand patients’ responses to DED and ascertain current limitations in the field of DED management, knowledge, attitudes, and practices (KAP) study is urgently warranted. KAP study is defined as a research methodology that systematically assesses individuals’ understanding, beliefs and behaviors related to a specific health issue [15]. Herein, knowledge refers to the awareness of DED causes, symptoms, risk factors, and available treatment options. Attitude encompasses patients’ feelings about the impacts of DED, as well as their beliefs towards preventive measures and treatment. Besides, practice includes adherence to recommended treatments, preventive measures, and lifestyle modifications. This study aimed to investigate the KAP among DED patients toward disease management and to provide guidance for clinical interventions. The findings can inform the development of targeted interventions to improve patient quality of life. Methods Study design and participants This cross-sectional survey was conducted between February 2023 and August 2024 at our hospital. The study participants were dry eye patients. The study was carried out after the protocol was approved by the Ethics Committee of Huai’an 82 Hospital (Approval No. 20230712001), and all participants provided written informed consent. The inclusion criteria were: (1) aged 18 years and older, (2) diagnosed with dry eye disease according to the Tear Film & Ocular Surface Society Dry Eye Workshop (TFOS-DEWS) definition [16, 17], based on typical symptoms and objective examinations including tear film break-up time (TBUT), Schirmer I test, and corneal/conjunctival fluorescein staining. (3) willing to participate in the study and provide informed consent. The exclusion criteria were: (1) self-reported participation in similar research projects, (2) presence of other significant ocular conditions such as corneal diseases, glaucoma, or retinal diseases that may impact dry eye symptoms or treatment preferences. The severity of dry eye disease was classified into mild, moderate, and severe according to the visual analog scale (VAS) score reported by the patients, with scores of 0–3 defined as mild, 4–6 as moderate, and 7–10 as severe. Questionnaire The questionnaire was developed with published literatures and guidelines [18–21]. The initial draft underwent revisions based on feedback from two senior experts specializing in ophthalmic diseases, both holding the title of chief physician. Subsequently, a preliminary trial was conducted on a limited scale (n = 52), resulting in a Cronbach’s alpha coefficient value of 0.912, indicating good internal consistency. The final questionnaire was in Chinese and consisted of four dimensions: demographic information, knowledge, attitudes and practices. The demographic information section contained 15 items. The knowledge, attitude, and practice dimensions comprised 11, 11, and 9 items, respectively. Among the demographic variables, “other eye diseases” referred to conditions such as cataract, glaucoma, or retinal disorders reported in addition to dry eye disease. In China, medical insurance typically covers outpatient visits, part of the costs of examinations, and common medications such as artificial tears, while coverage of newer or non-standard treatments may be limited. Therefore, the number and type of insurance held by participants could influence their access to dry eye care and treatment options. The knowledge items were scored 1 point for a correct answer and 0 points for others, resulting in a possible score range of 0–11 points. These items covered key aspects of dry eye disease, including common causes and risk factors (e.g., prolonged screen use, environmental conditions), typical symptoms (e.g., ocular dryness, visual disturbance), and treatment-related knowledge (e.g., artificial tears, anti-inflammatory medication, and the fact that treatment varies by disease severity).The attitude items scored on a five-point Likert scale ranging from very positive (5 points) to very negative (1 point), which is denoted as “P” or “N”, with a possible score range of 11–55 points. The practice items also scored on a five-point Likert scale, ranging from “Strongly agree” (5 points) to “Strongly disagree” (1 point), with a possible score range of 9–45 points. For example, attitude items included statements such as “I believe dry eye disease can significantly affect quality of life”, while practice items included statements such as “I actively seek knowledge about dry eye disease prevention and management from books or online resources.” The scoring rules for all dimensions are described in detail above to ensure clarity. The data were collected using an online questionnaire hosted on Sojump (http://www.sojump.com). At the beginning of the online questionnaire, the participants had to click the option of “I agree to participate this study” before they could answer the questions. All data were collected anonymously. To prevent duplication, an IP restriction was implemented, ensuring that each survey could only be completed once per unique IP address. Statistical analysis STATA 17.0 (Stata Corporation, College Station, TX, USA) was used for statistical analysis. The continuous variables were expressed as mean ± SD, and the categorical variables were expressed as n (%). The continuous variables conformed to a normal distribution were tested by the t-test or ANOVA. Normality of continuous variables was checked using the Shapiro–Wilk test, and homogeneity of variances was assessed before applying t-tests or ANOVA. In multivariate analysis, the cut-off value used in the study is 70% of the score distribution [22], with a cutoff of 7 points for knowledge, 41 points for attitudes, and 34 points for practices. Pearson’s correlation was used to analyze the correlation between knowledge, attitudes, and practices. The structural equation model (SEM) of KAP among dry eye patients toward disease management was constructed with AMOS 24.0 (IBM, NY, United States). The hypotheses as following: (1) knowledge had direct effects on attitude, (2) knowledge had direct effects on practice, and (3) attitude had direct effects on practice. The model fitting was evaluated with CMIN/DF (Chi-square fit statistics/degree of freedom), RMSEA (root mean square error of approximation), IFI (Incremental Fit Index), TLI (Tucker-Lewis Index), and CFI (Comparative Fit Index). Two-sided P < 0.05 were considered statistically significant. Results Characteristics of the participants Initially, 603 questionnaires were collected. Of these, 51 were excluded due to response times under 60 s, 7 due to respondents being under 18 years old, 2 for outlier values, and 1 for inconsistent responses regarding insurance status. Additionally, 42 questionnaires were discarded because participants selected “unclear” for all knowledge-related questions. As a result, 500 valid questionnaires were collected. Among the DED patients, 58.60% (293) were male, 26.80% (134) aged 43–54 years old and 62.00% (310) resided in urban areas. Lipid abnormality affected 25.00% (125) of participants, followed by mucin abnormality (15.80%; 79) and aqueous dynamic abnormality (12.20%; 61). Furthermore, 51.20% (256) had moderate symptoms of DED, and 54.00% (270) reached 3–5 points in the degree of eye discomfort. The mean scores for KAP were 5.80 ± 2.39 (possible range: 0–11), 38.68 ± 5.26 (possible range: 11–55), and 29.60 ± 8.22 (possible range: 9–45), respectively (Table 1). Table 1Demographic characteristics and KAP scoresVariablesN (%)Knowledge (mean ± SD) P Attitude (mean ± SD) P Practice (mean ± SD) P Total5005.80 ± 2.3938.68 ± 5.2629.60 ± 8.22 Gender 0.1730.0050.788 Male293 (58.60)5.92 ± 2.5339.24 ± 5.4129.52 ± 8.51 Female207 (41.40)5.63 ± 2.1837.89 ± 5.0229.72 ± 7.79 Age 0.0030.0010.004 ≤ 29 years old126 (25.20)6.26 ± 2.3339.52 ± 5.3831.30 ± 7.93 30–42 years old129 (25.80)6.10 ± 2.5139.63 ± 4.9028.228.78 43–54 years old134 (26.80)5.50 ± 2.5738.10 ± 5.6128.48 ± 8.82 ≥ 55 years old111 (22.20)5.30 ± 1.9737.33 ± 4.8930.64 ± 6.52 Residence 0.0950.0550.002 Rural190 (38.00)5.57 ± 2.2038.10 ± 5.0931.07 ± 6.31 Urban310 (62.00)5.94 ± 2.5039.04 ± 5.3928.70 ± 9.08 Marital status < 0.001< 0.0010.064 Unmarried96 (19.20)6.51 ± 2.2340.08 ± 5.6330.72 ± 8.94 Married323 (64.60)5.77 ± 2.5138.76 ± 5.1728.97 ± 8.63 Divorced or widowed81 (16.20)5.10 ± 1.8636.72 ± 4.7830.81 ± 4.65 Education < 0.001< 0.0010.007 Junior high school or below55 (11.00)5.15 ± 2.1637.56 ± 5.7929.64 ± 7.34 High school / technical school215 (43.00)5.27 ± 1.9637.42 ± 4.5630.95 ± 6.90 College and undergraduate207 (41.40)6.44 ± 2.6439.91 ± 5.5328.18 ± 9.25 Graduate and above23 (4.60)6.65 ± 2.7142.04 ± 4.3029.70 ± 9.84 Monthly per capita income < 0.001< 0.001< 0.001 < 2,00042 (8.40)5.60 ± 2.1138.57 ± 5.6530.76 ± 7.83 2,000–5,000240 (48.00)5.42 ± 2.0337.70 ± 4.7930.98 ± 6.71 5,000–10,000162 (32.40)6.06 ± 2.6839.24 ± 5.4928.36 ± 8.90 > 10,00056 (11.20)6.86 ± 2.8141.32 ± 5.4926.45 ± 10.70 Number of insurance 0.0450.0370.056 One type444 (88.80)5.71 ± 2.3638.49 ± 5.1729.91 ± 7.88 Two types and above46 (9.20)6.63 ± 2.5539.74 ± 5.6726.91 ± 9.15 No insurance10 (2.00)6.00 ± 2.8742.10 ± 7.2928.40 ± 14.99 Wearing contact lenses 0.0150.0010.322 Yes91 (18.20)5.25 ± 2.0137.07 ± 5.2230.37 ± 6.86 No409 (81.80)5.92 ± 2.4639.04 ± 5.2429.43 ± 8.48 Type of dry eye 0.004< 0.001< 0.001 Aqueous deficiency58 (11.60)6.05 ± 2.6539.38 ± 6.6131.29 ± 7.77 Lipid abnormality125 (25.00)5.55 ± 2.0737.42 ± 4.6532.17 ± 4.78 Mucin abnormality79 (15.80)5.30 ± 1.5637.24 ± 4.5430.99 ± 5.62 Aqueous dynamic abnormality61 (12.20)5.23 ± 1.9536.95 ± 5.0931.28 ± 4.63 Mixed type dry eye52 (10.40)6.40 ± 2.2939.87 ± 6.0229.79 ± 8.17 Don’t know125 (25.00)6.28 ± 3.0540.87 ± 4.5224.48 ± 11.23 Severity < 0.001< 0.001< 0.001 Mild203 (40.60)6.56 ± 2.7641.03 ± 4.8027.08 ± 10.53 Moderate256 (51.20)5.27 ± 1.9337.02 ± 4.8531.38 ± 5.67 Severe41 (8.20)5.37 ± 2.1537.37 ± 5.9031.00 ± 4.71 Frequency of eye discomfort in the past week 0.070< 0.0010.001 Never51 (10.20)6.20 ± 3.4741.39 ± 4.7425.69 ± 11.96 Occasionally219 (43.80)6.00 ± 2.3839.07 ± 4.8529.50 ± 8.53 Half of the time151 (30.20)5.36 ± 1.9637.46 ± 5.0331.22 ± 5.87 Most of the time70 (14.00)5.90 ± 2.3138.57 ± 6.2228.97 ± 7.63 All the time9 (1.80)5.22 ± 1.9935.11 ± 7.9132.11 ± 5.06 Degree of eye discomfort* 0.021< 0.001< 0.001 0 ~ 2 points112 (22.40)6.38 ± 3.0641.13 ± 4.6826.44 ± 11.50 3 ~ 5 points270 (54.00)5.72 ± 2.1538.13 ± 5.2630.39 ± 7.31 6 ~ 8 points108 (21.60)5.43 ± 2.1837.76 ± 5.1030.88 ± 5.18 9 ~ 10 points10 (2.00)5.50 ± 1.7236.10 ± 6.6729.90 ± 5.43 Other eye diseases 0.9150.0950.011 Yes102 (20.40)5.81 ± 2.3639.52 ± 5.2630.77 ± 8.87 No344 (68.80)5.78 ± 2.3538.34 ± 5.2129.72 ± 7.53 Don’t know54 (10.80)5.93 ± 2.7839.28 ± 5.7226.67 ± 10.31 Normal meibomian gland function 0.0240.001< 0.001 Yes171 (34.20)6.20 ± 2.5339.18 ± 5.6729.13 ± 8.59 No271 (54.20)5.56 ± 2.0537.98 ± 5.0031.57 ± 6.16 Don’t know58 (11.60)5.76 ± 3.2540.48 ± 4.9221.81 ± 10.49 Underwent eye surgery 0.9160.9950.001 Yes124 (24.80)5.78 ± 2.2938.68 ± 5.4531.70 ± 7.12 No376 (75.20)5.81 ± 2.4338.68 ± 5.2428.91 ± 8.43*Degree of Eye Discomfort: In this study, visual analog scale (VAS) Score was adapted to assess degree of eye discomfort, ranging from 0 to 10, with higher scores indicating more severe symptoms. “0” denotes the absence of a particular symptom, while “10” signifies a high severity of symptoms Demographic characteristics and KAP scores Frequency of eye discomfort in the past week *Degree of Eye Discomfort: In this study, visual analog scale (VAS) Score was adapted to assess degree of eye discomfort, ranging from 0 to 10, with higher scores indicating more severe symptoms. “0” denotes the absence of a particular symptom, while “10” signifies a high severity of symptoms Knowledge Knowledge scores varied significantly across age (P = 0.003), marital status (P < 0.001), education levels (P < 0.001), monthly per capita income (P < 0.001), number of insurance (P = 0.045), contact lens use (P = 0.015), dry eye type (P = 0.004), dry eye severity (P < 0.001), degree of eye discomfort (P = 0.021), and normal meibomian gland function (P = 0.024) (Table 1). The accuracy rates varied from 33.00% to 37.20%. For instance, 67.20% correctly recognized DED as the predominant ocular surface disorder, marked by reduced tear film stability and altered ocular surface balance (K1). However, only 33.00% accurately understood that treatment approaches do not remain uniform across all severity levels of dry eye disease (K11). Additionally, merely 35.00% were aware that dry eye disease involves more than simple “ocular dehydration” and its management extends beyond merely using eye drops (Table S1). Higher knowledge scores were positively associated with more favorable attitudes, indicating that greater awareness of dry eye disease tended to foster more proactive perspectives toward its management. Attitude Significant differences were observed among genders (P = 0.005), age (P = 0.001), marital status (P < 0.001), education levels (P < 0.001), monthly per capita income (P < 0.001), number of insurance (P = 0.037), contact lens use (P = 0.001), dry eye type (P < 0.001), dry eye severity (P < 0.001), frequency of eye discomfort in the past week (P < 0.001), degree of eye discomfort (P < 0.001), and normal meibomian gland function (P = 0.001) (Table 1). The majority (69.2%) expressed a positive attitude toward the importance of actively communicating with their attending physician, and seeking advice on disease prevention and management related to dry eye disease (A3). However, only 10.8% disagreed with the notion that the financial stress associated with the treatment of dry eye disease could lead to anxiety or a feeling of melancholy (A4) (Table S2). Attitude scores were positively linked with practice, suggesting that patients who held more positive views were more likely to engage in active management behaviors. Practice Practice scores exhibited significant differences across age (P = 0.004), residences (P = 0.002), education levels (P = 0.007), monthly per capita income (P < 0.001), dry eye type (P < 0.001), dry eye severity (P < 0.001), frequency of eye discomfort in the past week (P = 0.001), degree of eye discomfort (P < 0.001), presence of other eye diseases (P = 0.011), normal meibomian gland function (P < 0.001), and a history of eye surgery (P = 0.001) (Table 1). Participants who reported other ocular conditions (e.g., cataract or glaucoma) exhibited minor variations in practice scores; nevertheless, as these comorbidities were outside the primary scope of the present study, detailed interpretation was not further pursued. About 61.2% demonstrated adherence by actively seeking knowledge related to dry eye prevention and management through various channels, such as books, the internet, and newspapers (P1). However, only 45.4% engaged in actively participating in lectures on dry eye disease or in-hospital education sessions conducted by community hospitals or ophthalmology departments (P2) (Table S3). Correlation analysis and multivariate logistic analysis Pearson’s correlation analysis indicated positive association between knowledge and attitudes (r = 0.482, P < 0.001) (Table S4). Multivariate logistic analysis demonstrated that being divorced or widowed (OR = 0.382, 95% CI: 0.165–0.886, P = 0.025) and moderate dry eye severity (OR = 0.459, 95% CI: 0.269–0.781, P = 0.004) were independently associated with knowledge (Table 2). Knowledge (OR = 1.456, 95% CI: 1.308–1.621, P < 0.001), moderate dry eye severity (OR = 0.452, 95% CI: 0.254–0.804, P = 0.007) and occasionally experiencing eye discomfort in the past week (OR = 0.431, 95% CI: 0.189–0.982, P = 0.045) were independently associated with attitudes (Table 3). Attitudes (OR = 1.082, 95% CI: 1.033–1.134, P = 0.001), being divorced or widowed (OR = 0.447, 95% CI: 0.211–0.944, P = 0.035), aqueous deficiency (OR = 2.103, 95% CI: 1.036–4.272, P = 0.040) and lipid abnormality (OR = 2.937, 95% CI: 1.616–5.338, P < 0.001) were independently associated with practices (Table 4). Table 2Univariate and multivariate logistic regression analysis of knowledgeUnivariate logistic regressionMultivariate logistic regressionOR (95%CI) P OR (95%CI) P Gender Male1.468 (1.005–2.145)0.0471.058 (0.683–1.639)0.800 Femalerefref Age ≤ 29 years old1.975 (1.150–3.391)0.0141.193 (0.597–2.384)0.618 30–42 years old1.512 (0.878–2.603)0.1360.918 (0.498–1.690)0.783 43–54 years old0.977 (0.560–1.705)0.9350.713 (0.385–1.321)0.282 ≥ 55 years oldrefref Residence Rural0.783 (0.534–1.148)0.210 Urbanref Marital status Unmarriedrefref Married0.524 (0.330–0.830)0.0060.800 (0.419–1.528)0.500 Divorced or widowed0.246 (0.125–0.485)< 0.0010.382 (0.165–0.886)0.025 Education Junior high school or below0.256 (0.091–0.723)0.0100.358 (0.100-1.285)0.115 High school / Technical school0.315 (0.132–0.755)0.0100.437 (0.139–1.373)0.156 College and undergraduate0.793 (0.335–1.878)0.5980.833 (0.285–2.431)0.738 Graduate and aboverefref Monthly per capita income < 2,0000.496 (0.220–1.122)0.0921.423 (0.511–3.961)0.500 2,000–5,0000.281 (0.154–0.512)< 0.0010.699 (0.316–1.545)0.377 5,000–10,0000.554 (0.300-1.023)0.0590.973 (0.450–2.102)0.944 > 10,000ref Number of insurance One type0.323 (0.090–1.163)0.084 Two types and above0.727 (0.181–2.923)0.654 No insuranceref Wearing contact lenses Yes0.572 (0.342–0.956)0.0330.581 (0.327–1.034)0.065 Norefref Type of dry eye Aqueous deficiency0.765 (0.408–1.435)0.4040.941 (0.443–1.996)0.873 Lipid abnormality0.389 (0.229–0.660)< 0.0010.769 (0.399–1.482)0.432 Mucin abnormality0.320 (0.170–0.601)< 0.0010.665 (0.317–1.395)0.281 Aqueous dynamic abnormality0.385 (0.197–0.753)0.0050.681 (0.312–1.487)0.334 Mixed type dry eye0.929 (0.486–1.776)0.8231.164 (0.537–2.523)0.701 Don’t knowrefref Severity of dry eye Mildrefref Moderate0.316 (0.213–0.471)< 0.0010.459 (0.269–0.781)0.004 Severe0.363 (0.173–0.764)0.0080.666 (0.235–1.887)0.444 Frequency of eye discomfort in the past week Neverrefref Occasionally0.685 (0.372–1.264)0.2271.266 (0.604–2.652)0.532 Half of the time0.350 (0.181–0.677)0.0020.901 (0.389–2.084)0.807 Most of the time0.543 (0.259–1.135)0.1051.210 (0.467–3.137)0.695 All the time0.130 (0.015–1.116)0.0630.387 (0.031–4.781)0.459 Degree of eye discomfort 0 ~ 2 pointsrefref 3 ~ 5 points0.428 (0.273–0.673)< 0.0010.769 (0.426–1.387)0.382 6 ~ 8 points0.443 (0.256–0.768)0.0040.992 (0.447-2.200)0.985 9 ~ 10 points0.107 (0.013–0.875)0.0370.206 (0.020–2.158)0.187 Other eye diseases Yesref No0.846 (0.534–1.339)0.475 Don’t know1.158 (0.589–2.274)0.670 Normal meibomian gland function Yesrefref No0.630 (0.422–0.941)0.0240.935 (0.585–1.495)0.779 Don’t know1.120 (0.613–2.046)0.7130.712 (0.350–1.450)0.349 Undergoing eye surgery Yesref No1.121 (0.730–1.722)0.602 Univariate and multivariate logistic regression analysis of knowledge Table 3Univariate and multivariate logistic regression analysis of attitudesUnivariate logistic regressionMultivariate logistic regressionOR (95%CI) P OR (95%CI) P Knowledge 1.531 (1.389–1.688)< 0.0011.456 (1.308–1.621)< 0.001 Gender Male1.668 (1.138–2.443)0.0091.315 (0.819–2.111)0.256 Femalerefref Age ≤ 29 years old1.239 (0.725–2.120)0.433 30–42 years old1.276 (0.749–2.175)0.370 43–54 years old1.019 (0.596–1.742)0.947 ≥ 55 years oldref Residence Rural0.772 (0.526–1.132)0.185 Urbanref Marital status Unmarriedrefref Married0.663 (0.418–1.053)0.0821.146 (0.626-2.100)0.658 Divorced or widowed0.404 (0.212–0.771)0.0061.053 (0.466–2.378)0.902 Education Junior high school or below0.344 (0.126–0.939)0.0370.498 (0.122–2.032)0.331 High school / Technical school0.264 (0.110–0.637)0.0030.420 (0.117–1.510)0.184 College and undergraduate0.603 (0.253–1.439)0.2550.450 (0.133–1.524)0.200 Graduate and aboverefref Monthly per capita income < 2,0000.605 (0.270–1.355)0.2221.992 (0.649–6.116)0.229 2,000–5,0000.269 (0.147–0.491)< 0.0010.841 (0.347–2.039)0.701 5,000–10,0000.569 (0.308–1.050)0.0711.141 (0.481–2.708)0.765 > 10,000refref Number of insurance One type0.212 (0.054–0.832)0.0260.307 (0.051–1.846)0.197 Two types and above0.393 (0.090–1.710)0.2130.289 (0.042–1.978)0.206 No insurancerefref Wearing contact lenses Yes0.424 (0.246–0.729)0.0020.536 (0.281–1.023)0.059 Norefref Type of dry eye Aqueous deficiency0.652 (0.347–1.223)0.1820.859 (0.368–2.001)0.724 Lipid abnormality0.291 (0.170–0.500)< 0.0010.552 (0.264–1.153)0.114 Mucin abnormality0.253 (0.133–0.481)< 0.0010.697 (0.314–1.547)0.375 Aqueous dynamic abnormality0.386 (0.201–0.742)0.0040.935 (0.404–2.168)0.876 Mixed type dry eye0.677 (0.352-1.300)0.2410.802 (0.348–1.849)0.605 Don’t knowrefref Severity of dry eye Mildrefref Moderate0.241 (0.160–0.361)< 0.0010.452 (0.254–0.804)0.007 Severe0.408 (0.200-0.833)0.0140.945 (0.318–2.811)0.919 Frequency of eye discomfort in the past week Neverrefref Occasionally0.302 (0.160–0.571)< 0.0010.431 (0.189–0.982)0.045 Half of the time0.171 (0.086–0.339)< 0.0010.424 (0.170–1.060)0.066 Most of the time0.342 (0.162–0.725)0.0050.677 (0.237–1.933)0.466 All the time0.156 (0.029–0.830)0.0290.482 (0.056–4.183)0.508 Degree of eye discomfort 0 ~ 2 pointsrefref 3 ~ 5 points0.378 (0.240–0.595)< 0.0010.900 (0.470–1.725)0.752 6 ~ 8 points0.381 (0.219–0.663)0.0011.013 (0.427–2.403)0.977 9 ~ 10 points0.096 (0.012–0.786)0.0290.112 (0.008–1.496)0.098 Other eye diseases Yesrefref No0.620 (0.394–0.974)0.0380.630 (0.353–1.125)0.118 Don’t know0.906 (0.464–1.770)0.7730.556 (0.218–1.417)0.219 Normal meibomian gland function Yesrefref No0.666 (0.445–0.998)0.0491.108 (0.658–1.865)0.701 Don’t know1.705 (0.935–3.106)0.0821.667 (0.688–4.036)0.258 Undergoing eye surgery Yesref No1.031 (0.673–1.579)0.888 Univariate and multivariate logistic regression analysis of attitudes Table 4Univariate and multivariate logistic regression analysis of practicesUnivariate logistic regressionMultivariate logistic regressionOR (95%CI) P OR (95%CI) P Knowledge 1.018 (0.940–1.103)0.6580.935 (0.849–1.030)0.174 Attitudes 1.058 (1.019–1.098)0.0031.082 (1.033–1.134)0.001 Gender Male0.873 (0.593–1.284)0.491 Femaleref Age ≤ 29 years old1.325 (0.764–2.297)0.317 30–42 years old0.993 (0.567–1.739)0.980 43–54 years old0.977 (0.560–1.705)0.935 ≥ 55 years oldref Residence Rural0.946 (0.638–1.403)0.782 Urbanref Marital status Unmarriedrefref Married0.894 (0.552–1.447)0.6480.822 (0.490–1.377)0.456 Divorced or widowed0.434 (0.215–0.875)0.0200.447 (0.211–0.944)0.035 Education Junior high school or below0.531 (0.189–1.494)0.230 High school / Technical school0.851 (0.352–2.056)0.719 College and undergraduate0.522 (0.213–1.276)0.154 Graduate and aboveref Monthly per capita income < 2,0001.846 (0.775–4.398)0.166 2,000–5,0001.390 (0.716–2.698)0.330 5,000–10,0001.154 (0.575–2.314)0.687 > 10,000ref Number of insurance One type0.648 (0.180–2.335)0.507 Two types and above0.591 (0.143–2.441)0.467 No insuranceref Wearing contact lenses Yes1.033 (0.631–1.692)0.896 Noref Type of dry eye Aqueous deficiency1.935 (0.990–3.784)0.0542.103 (1.036–4.272)0.040 Lipid abnormality2.182 (1.267–3.759)0.0052.937 (1.616–5.338)< 0.001 Mucin abnormality1.073 (0.559–2.061)0.8311.495 (0.748–2.988)0.255 Aqueous dynamic abnormality0.943 (0.457–1.946)0.8741.380 (0.640–2.974)0.411 Mixed type dry eye1.167 (0.558–2.439)0.6821.221 (0.571–2.607)0.607 Don’t knowrefref Severity of dry eye Mildref Moderate1.229 (0.821–1.838)0.316 Severe0.939 (0.441–1.999)0.871 Frequency of eye discomfort in the past week Neverref Occasionally1.104 (0.567–2.150)0.771 Half of the time1.019 (0.508–2.044)0.958 Most of the time0.894 (0.402–1.990)0.784 All the time1.200 (0.265–5.438)0.813 Degree of eye discomfort 0 ~ 2 pointsref 3 ~ 5 points0.937 (0.584–1.504)0.788 6 ~ 8 points0.812 (0.455–1.448)0.480 9 ~ 10 points0.528 (0.107–2.613)0.434 Other eye diseases Yesrefref No0.617 (0.389–0.979)0.0410.726 (0.446–1.181)0.197 Don’t know0.491 (0.235–1.030)0.0600.692 (0.316–1.518)0.359 Normal meibomian gland function Yesref No1.424 (0.937–2.164)0.098 Don’t know0.485 (0.221–1.064)0.071 Undergoing eye surgery Yesref No0.925 (0.596–1.434)0.726 Univariate and multivariate logistic regression analysis of practices SEM The fitting indices of the structural equation model (SEM) outperformed the respective threshold value (CMIN/DF = 2.484, RMSEA = 0.055, IFI = 0.877, TLI = 0.867, CFI = 0.877), indicating satisfactory model fit (Table S5). In the SEM, direct effects were observed between knowledge and attitudes (path coefficient = 2.950, P < 0.001), knowledge and practice (path coefficient=-2.474, P = 0.002), and attitudes and practices (path coefficient = 0.595, P < 0.001) (Fig. 1 and Table S6). Figure 1 illustrates the structural equation model of knowledge, attitudes, and practices, showing the standardized path coefficients between the latent variables. Knowledge had a positive effect on attitudes (0.71), attitudes positively influenced practices (0.42), while knowledge showed a negative direct effect on practices (–0.42). Structural equation model Discussion DED patients demonstrated insufficient knowledge, passive attitudes, and poor practices toward DED in this study. These findings carried significant implications for healthcare interventions and educational programs aimed at enhancing the KAP levels of DED patients concerning the condition. Although dry eye disease is generally reported to be more prevalent among females and older individuals in epidemiological studies, our sample contained a higher proportion of male and relatively younger participants. This discrepancy may be related to the hospital-based sampling and the online recruitment method, in which male patients and younger individuals with more frequent screen use were more likely to participate. In addition, regional and demographic characteristics of the participants might have contributed to this distribution. While these differences in gender and age composition do not alter the main findings regarding the associations among knowledge, attitudes, and practices, they highlight the need for future studies to include more diverse and representative populations. In the knowledge dimension, Item K11 had the lowest correctness rate of 33.00%, indicating lack of knowledge about the treatment regimens across varying severity of dry eye. This finding highlighted a knowledge gap in the appropriate treatment approaches for dry eye disease, potentially leading to suboptimal management strategies based on severity levels. This discrepancy could be attributed to the evolving nature of dry eye management, with new treatments and research emerging [23, 24], or it might reflect the complexity of tailoring treatments to individual patient needs. Similarly, Item K5 had a correctness rate of only 35.00%, suggesting a misunderstanding of DED beyond being characterized as “ocular dehydration” and the misconception that its management relies solely on the application of eye drops. It could indicate a need for improved education and awareness about the diverse causes and management strategies for DED [25]. In the attitude dimension, merely 19.2% agreed that DED, being a relatively common ocular condition, could result in severe adverse consequences and should receive adequate attention. This finding highlighted a discrepancy in the perception of the severity and significance of DED among the participants. Addressing this disparity is crucial, as DED, despite its commonality, can have a significant impact on visual function and ocular health if left unmanaged [26, 27]. In the practice dimension, 46.2% actively sought to implement recommended home-based techniques, such as eye massages and steaming, to alleviate dry eye disease symptoms. Healthcare providers should consider enhancing the accessibility and attractiveness of educational programs on evidence-based home-care strategies to encourage greater participation and dissemination of accurate information to supplement clinical treatments effectively [28, 29]. The positive associations between knowledge and attitude scores, as well as between attitude and practice scores towards DED, aligned with health behavior theories [30]. Individuals with good knowledge of DED are more likely to have positive attitudes, which, in turn, translate into practices for the prevention and management of the condition. However, negative association between knowledge and practice scores from SEM models warrants further exploration. This phenomenon was explained from multiple perspectives, including individual cognitive factors, cultural beliefs, healthcare system barriers, and economic burden. One possible explanation is that, patients with knowledge of DED are more acutely aware of the intricacy of treatment, and then have feelings of bewilderment and confusion, thereby resulting in reduced engagement in treatment [31]. Beyond individual cognitive factors, cultural beliefs, limited accessibility of specialized care, and financial burden may also contribute to this counterintuitive result. In the Chinese healthcare context, patients may have knowledge about DED management but face barriers such as restricted access to ophthalmology services, limited insurance coverage for advanced treatments, and treatment-related costs. These factors could prevent knowledge from being effectively translated into practice. Greater awareness of the complexity of DED treatment may cause patients to hesitate or feel uncertain, which could reduce their willingness to apply what they know in daily management. To clarify this unexpected link, future studies may draw on qualitative approaches, such as focus groups or patient interviews, to capture patient perspectives more directly. Several influential factors of KAP scores were further identified. First, divorced or widowed individuals were less likely to have adequate knowledge or to engage in effective management for dry eye disease compared to single counterparts. Reduced social support and associated psychological stress may partly explain this pattern [32]. Previous studies have reported that marital disruption and social isolation can negatively influence health literacy, disease awareness, and adherence to medical advice, which may, in turn, impair patients’ ability to acquire and apply disease-related knowledge [32, 33]. Moreover, empirical evidence has shown that social support and marital status are significant predictors of health literacy and self-management behaviors among patients with chronic diseases, highlighting the importance of social relationships in shaping disease-related knowledge and behaviors [34].Second, patients with moderate dry eye severity possess less knowledge and less positive attitudes towards their condition. Reportedly, increased severity in chronic conditions can lead to decreased quality of life and greater emotional distress, thereby impacting patient engagement and educational outcomes [33]. Moderate conditions are also less likely to trigger the more intensive educational or therapeutic interventions reserved for severe cases, potentially leading to a gap in patient knowledge and attitude development. Third, occasional eye discomfort over the past week demonstrated a significant negative association with their attitudes compared to those without DED. The finding can be explained by unpredictable discomfort and irregularity of symptoms, which may foster feelings of frustration or anxiety about the condition [35]. Fourth, both aqueous deficiency and lipid abnormality subtypes of DED were significantly associated with more proactive practices. Aqueous deficiency, characterized by decreased tear production, often leads to more severe and persistent symptoms [36]. Similarly, lipid abnormalities in the tear film can lead to increased evaporation of tears, and cause significant discomfort [37]. The DED subtypes can thus motivate patients to follow treatment plans rigorously to alleviate symptoms. The findings have implication for healthcare providers to consider the subtype of DED when advising patients, as different subtypes may require distinct management strategies and patient education approaches. These results suggest that identifying the DED subtype may help clinicians tailor patient education and reinforce adherence strategies, such as emphasizing artificial tear use for aqueous deficiency or lipid-based therapy and lid hygiene for lipid abnormality. Several limitations should be acknowledged. Firstly, the cross-sectional design and limited sample size preclude causal inference between knowledge, attitudes, and practices and may constrain the generalizability of the findings. Secondly, reliance on self-reported questionnaires could introduce recall or social desirability bias, potentially leading to score inflation [38]. Although these biases are inherent in self-reported surveys, they may limit the precision of our findings. Future studies could consider combining self-reports with objective measures, such as digital monitoring tools, electronic health records, or audits of medical records, to enhance the accuracy and robustness of the results. Additionally, as the questionnaire was originally designed and validated in Chinese, its applicability to populations from different linguistic and cultural backgrounds may be limited. Future studies should consider adapting and validating this instrument in diverse settings to enhance the generalizability of the findings. Thirdly, since the number of distributed questionnaires was not recorded, the response rate could not be assessed, and potential response bias cannot be excluded. As the questionnaire was distributed online through a QR code or link via Questionnaire Star and shared multiple times across different platforms, it was difficult to track how many individuals actually viewed the questionnaire. In addition, because the survey was conducted online, patients without internet access—particularly those in rural areas—may have been underrepresented, which could limit the representativeness of the sample. Moreover, although DED was diagnosed according to TFOS-DEWS criteria and based on commonly used tests (tear film break-up time, Schirmer I test, and corneal/conjunctival fluorescein staining), other diagnostic indices such as tear osmolarity or MMP-9 were not included due to limited resources, which may have restricted the comprehensiveness of case classification. Furthermore, screen usage intensity, which has been reported as an important factor associated with dry eye disease, was not assessed in this study, potentially limiting the ability to capture practice-related influences. The demographic profile of our sample, which included a relatively higher proportion of younger and male patients, may also differ from the typical distribution reported in epidemiological studies; this likely reflects the characteristics of hospital-based and online recruitment. Finally, information on disease duration, healthcare facility types, and consultation frequency was not collected, which may have further constrained interpretation. Nevertheless, unlike most previous studies that primarily examined KAP among healthcare professionals, this study provides novel insights into the perspectives of DED patients, thereby offering valuable guidance for patient-centered education and clinical care. Conclusion In conclusion, DED patients exhibited limited knowledge, suboptimal attitudes, and inadequate practices regarding disease management. These findings underscore the urgent need for tailored educational and training interventions to improve patient engagement, optimize adherence to treatment, and ultimately enhance clinical outcomes. For example, targeted interventions tailored to different DED subtypes, community-based educational workshops, and online platforms could be implemented to provide patients with accessible and practical resources for disease management. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 2 Acknowledgements Not applicable. Abbreviations Knowledge, attitudes and practices Dry eye disease Tear Film & Ocular Surface Society Dry Eye Workshop Structural equation model Chi-square fit statistics/degree of freedom Root mean square error of approximation Incremental fit Index) Tucker-Lewis index Comparative fit index Author contributions Hongwei Zhou carried out the studies, participated in collecting data, and drafted the manuscript. Chongyan Zhu and Hongwei Zhou performed the statistical analysis and participated in its design. Hongwei Zhou and Yajie Zhou participated in acquisition, analysis, or interpretation of data and draft the manuscript. All authors read and approved the final manuscript. Funding This study was supported by Hospital-level research project of Huai’an 82 hospital (2024YB10). Data availability All data generated or analyzed during this study are included in this article and supplementary information files. Declarations Ethics approval and consent to participate I confirm that all methods were performed in accordance with the relevant guidelines. This work has been carried out in accordance with the Declaration of Helsinki (2000) of the World Medical Association. The study was carried out after the protocol was approved by the Ethics Committee of Huai’an 82 Hospital (Approval No. 20230712001), and all participants provided written informed consent. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Contributor Information Yajie Zhou, Email: Zhouyajie_ha@163.com. Hongwei Zhou, Email: leave0307@163.com. References Associated Data Supplementary Materials Supplementary Material 1 Supplementary Material 2 Data Availability Statement All data generated or analyzed during this study are included in this article and supplementary information files.