Persons contacted and examined among a cohort of 313 patients operated on for cataract at Sanrao Hospital in Sanrao, China.
Congdon NG, Rao SK, Zhao X, Wang W, Choi K, Lam DSC. Visual Function and Postoperative Care After Cataract Surgery in Rural ChinaStudy of Cataract Outcomes and Up-Take of Services (SCOUTS) in the Caring Is Hip Project, Report 2. Arch Ophthalmol. 2007;125(11):1546-1552. doi:10.1001/archopht.125.11.eeb70014
To study the postoperative visual function and uptake of refraction and second-eye surgery among persons undergoing cataract surgery in rural China.
Self-reported visual function was measured 10 to 14 months after surgery. Subjects with improvement of 2 or more lines with refraction were offered glasses, and those with significant cataract were offered second-eye surgery.
Among 313 eligible subjects, 242 (77%) could be contacted; 176 (73%) of those contacted were examined. Interviewed subjects had a mean ± SD age of 69.9 ± 10.2 years, and 63.6% were female. The mean ± SD visual function score was 88.4 ± 12.3, higher than previously reported for cataract programs in rural China and significantly (P = .03) correlated with presenting vision. Forty-two percent of subjects had spectacles, more than half being reading glasses. Though 87% of subjects' vision improved with refraction, only 35% accepted prescriptions, the most common reason for refusal being lack of perceived need. Second-eye surgery was accepted by a total of 48% (85 of 176) of patients, cost being the biggest reason for refusal.
Visual function was high in this cohort. Potential benefit of refraction and second-eye surgery was substantial, but uptake of services was modest. Programs to improve service uptake should focus on reading glasses and cost-reduction strategies such as tiered pricing.Published online October 22, 2007 (doi:10.1001/archophthalmol.125.11.eeb70014).
Poor visual outcomes,1- 6 low visual function (VF) and quality of life,1,2,6 and poor uptake of services7 have been reported in conjunction with cataract surgery in rural Asia. These undesirable results are particularly striking when compared with the often excellent outcomes reported for urban centers in India8,9 and China.10 Studies attempting to pinpoint the reasons for poor cataract surgical outcomes in rural Asia have identified underutilization of postoperative refractive services as a leading factor in India,3- 5 Bangladesh,11 Pakistan,12 and China.2
Few studies in rural Asia have reported in detail the impact of postoperative interventions such as refraction on visual outcomes of cataract surgery. Even fewer have assessed the potential for uptake of such services. Recent research in rural Africa has suggested a large unmet need for presbyopic correction,13 but this problem remains largely unstudied in rural Asia. The impact of patients' own perceptions of their VF on uptake of postoperative refractive services and second-eye surgery is also poorly understood in this setting.
The Sanrao Study of Cataract Outcomes and Up-Take of Services (SCOUTS) is an intensive study of outcomes and service uptake among approximately 300 persons undergoing sutureless, manual cataract extraction performed by local surgeons in rural China. The present report examines: (1) Patient self-reported VF after surgery. (2) Potential impact of second-eye surgery and postoperative refraction on near and distance vision after cataract surgery. (3) Patient uptake of refractive services and second-eye surgery and factors (including self-reported VF) that influence these decisions.
The methods for SCOUTS have been reported in detail elsewhere.14 Persons undergoing cataract surgery by either of 2 local, rural surgeons through Project Vision (the eye care component of the Caring is Hip medical relief program) at the Sanrao Hospital in rural Guangdong Province, China, between August 8 (the first date on which the surgeons performed independent surgery after completing training) and December 31, 2005, were invited by telephone approximately 1 year after surgery to return to the hospital for a comprehensive ocular examination and series of questionnaires relating to spectacle use and VF. Sanrao Hospital is a village-level, government-run facility. Written informed consent was obtained from all subjects, and the study protocol was approved by the institutional review board at the Joint Shantou International Eye Center, which serves as a parent hospital for Sanrao. The Declaration of Helsinki was followed in all study procedures.
All persons undergoing cataract surgery at Sanrao received a preoperative examination with measurement of visual acuity and dilation of the pupil. Data recorded for all patients included age, sex, preoperative presenting acuity, intraocular lens power, and presenting acuity on the first postoperative day.
Having had at least 1 eye operated on for cataract at Sanrao by the 2 recently trained study surgeons between the period of August 8 and December 31, 2005, was the only eligibility criterion for the study. All eligible patients seen at Sanrao Hospital between October 16 and 21, 2006 (10-14 months after surgery), underwent a study-specific examination including measurement of visual acuity in each eye separately at near and distance with and without refraction (KR-8800; Topcon Optical [H. K.] Limited, Hong Kong, China), with subjective refinement by an ophthalmologist. Distance visual acuity was measured using an illuminated tumbling E Snellen chart at a distance of 6 m, and near visual acuity was measured at 33 cm with a handheld chart. Subjects had to correctly state more than half of the optotypes on a line (eg, 3 of 5, 4 of 6) to proceed to the next line.
A single fellowship-trained glaucoma specialist (N.G.C.) evaluated the anterior segment by slitlamp biomicroscopy (YZ5F1; Suzhou Liuliu, Suzhou, China) and performed gonioscopy; subjects underwent dilation of the pupil in both eyes, after yttrium aluminum garnet (YAG) laser iridotomy, if thought to be indicated clinically. The same examining ophthalmologist evaluated the posterior segment by slitlamp biomicroscopy with the aid of a 90-diopter (D) lens and indirect ophthalmoscopy with a 20-D lens. Evidence of the following complications was recorded by the examiner: abnormalities of wound architecture, presence of an irregular pupil, iris adherent to the wound, vitreous visible in the anterior chamber, decentration of the intraocular lens, visible capsular rent, or cystoid macular edema.
Subjects whose near or distance visual acuity improved by 2 or more lines in either eye were offered near and/or distance prescriptions to be filled at a nearby optical shop, and their acceptance or refusal was recorded. Prior use of near and distance glasses was also recorded for all subjects. Subjects who were thought by the examiner to have visually significant cataract in the second eye were offered surgery under the same terms as for the first eye (a price of $90, inclusive of intraocular lens and postoperative care). Subjects refusing the prescription or surgery were asked to indicate their reason for refusal. Subjects were also offered YAG capsulotomy where thought to be clinically indicated; these results are reported elsewhere.15
The VF questionnaire was a Chinese translation of an instrument developed originally by Fletcher et al16 for use in rural Asia. All questions were administered in the local dialects (Chaoshanhua and Kejiahua) by 1 of 3 native speakers after a period of training and standardization. This instrument has previously been validated for use in Chinese1,2 and is described elsewhere in detail.16
Briefly, the VF questionnaire assesses overall vision, visual perception, limitation in daily activities, peripheral vision, near vision, sensory adaptation, light-dark adaptation, visual search, color discrimination, glare disability, and depth perception. The questionnaire could be administered in 5 to 10 minutes. Each of 13 responses was scored from 1 (no problems) through 4 (maximum problems), with scales in each of the areas calibrated between 100 (the best possible score) and 0 (the worst score). The overall VF scale score was calculated by averaging the scores for the different areas, thus giving a summary score of 0 to 100.16
Subjects who were unwilling or unable to return to the Sanrao Hospital for examination were requested to complete the questionnaire by telephone, in which case the instrument was administered by the identical 3 trained study personnel using the identical script as for subjects interviewed at the time of examination.
Univariate analyses were done using the t test, Mann-Whitney test, χ2 test, or Fisher exact test, as appropriate. Multiple linear regression modeling was used to examine potential predictors for VF. Logistic regression was used to assess potential factors associated with willingness to accept spectacle prescription and uptake of second-eye cataract surgery. Generalized estimating equations (GEE) models were used to assess the association between potential predictors and each of the outcome variables, thus accounting for intracorrelated data from subjects with bilateral surgery. Statistical analyses were done using SPSS 14.0 (SPSS Inc, Chicago, Illinois). All statistical tests were 2-sided, and a P value < .05 was considered statistically significant.
Among 313 persons operated on within the study window, 242 (77%) could be contacted by telephone; study examinations and interviews were performed on 176 (73%), 63 (26%) underwent telephone interviews without examination, and 3 (1%) refused examination or interview (Figure). All subjects had telephones in the house; reasons for noncontact were failure to answer or changes in the telephone number in the time since surgery. Thus, 239 of 313 eligible subjects (76%) completed the VF interview (Table 1). These 239 subjects had a mean ± SD age of 69.9 ± 10.2 years, 36.4% (87 of 239) were male, and 87.0% (208 of 239) had been blind (presenting vision ≤ 6/60) in the first eye operated on prior to surgery.
Reading was uncommon among these rural Chinese subjects: among 239 interviewed subjects, 10.7% (n = 25) read once a week or more and 74.5% (n = 178) indicated that they did not read at all (Table 1). Examined subjects and those undergoing only telephone interviews did not differ significantly in important characteristics (Table 1). As has been reported previously,14 examined and interviewed patients did not differ significantly from those who could not be contacted with regard to age, sex, preoperative presenting visual acuity, or day 1 postoperative presenting visual acuity in the eye operated on.
Among 176 examined patients, 109 had only one eye operated on by the 2 Sanrao surgeons, 32 had one eye operated on by the 2 Sanrao surgeons and the fellow eye was operated on by others (5 fellow eyes were operated on by Sanrao surgeons not participating in the study and 27 were operated on locally outside the Sanrao facility), and 35 were operated on in both eyes by the 2 Sanrao surgeons (Figure). Thus, a total of 211 eyes were operated on by the study surgeons (109 by surgeon A and 102 by surgeon B). As reported elsewhere,14 evidence of intraoperative or postoperative complications was present in 8.5% (18 of 211) of the eyes operated on by the 2 trained Sanrao surgeons, and 95.7% (202 of 211) of examined study eyes had best-corrected visual acuity of 6/18 or better.
The mean ± SD VF score for 239 persons undergoing interviews was 88.4 ± 12.3. These results are considerably better (higher scores) than reported for 2 other studies from rural China,1,2 which had high rates of postoperative blindness (presenting visual acuity ≤ 6/60). In Doumen County, China,1 where 52.6% of 109 subjects were blind postoperatively, the mean ± SD VF score using the same instrument16 was 41.6 ± 20.0, while in Shunyi County, China2 (postoperative blindness present among 44.8% of 87 subjects), the mean ± SD score was 61.9 ± 30.0. A study from urban India with similar low rates of postoperative blindness to ours among 1700 subjects (Aravind Eye Hospital, 1.1%; current study, 3.3%) reported a similar mean ± SD VF score of 79.8 ± 20.0.9
Potential predictors of VF score were examined in GEE linear regression models. In the final model, postoperative presenting distance visual acuity (P < .005) and bilateral (as compared with unilateral) surgery (P = .03) were significantly associated with VF (Table 2). Sex, age, surgeon, preoperative visual acuity, presence of operative complications, near visual acuity (in a separate model, not shown), and reading (ever vs never) were not significantly associated with VF.
Previous spectacle wear was not uncommon among 176 patients who underwent examination: 42.2% (n = 73) of subjects had previously been prescribed spectacles of some kind (Table 3). Among those with glasses, the most common were spectacles for reading (n = 41, 56.2% of those with glasses and 23.7% of all subjects), followed by plano glasses worn for protection (n = 14, 19.2% of those with glasses and 8.1% of all subjects) and distance spectacles (n = 11, 15.1% of those with glasses and 6.4% of all subjects) (Table 3). Reading spectacles were also used more regularly than other types of glasses, with 70% of respondents who wore reading glasses indicating that they were used regularly, as compared with only 38.5% of those with plano spectacles, though the difference was not statistically significant (Table 3) (P > .3).
A substantial proportion of subjects could benefit from refraction: among 174 examined subjects with complete refraction data, 151 (86.8%) could improve by 2 or more lines at distance and/or near over their presenting visual acuity in at least 1 eye (Table 4). Uptake of refractive services, however, was low: only 53 subjects (35.1%) with visual improvement on refraction indicated a willingness to receive a written prescription for spectacles. This proportion did not differ significantly between persons with an improvement in distance vision only, near vision only, or both (Table 4). Subjects refused glasses principally because of a lack of perceived need: 79% (77 of 98) of subjects refusing glasses indicated that they were satisfied with their current vision, as opposed to 7% (7 of 98) who feared the cost of glasses and 14% (14 of 98) who found spectacles uncomfortable.
In logistic regression models including potential factors predictive of willingness to accept spectacle prescriptions, the most important association with accepting a new prescription was having previously owned glasses (odds ratio, 10.97; 95% confidence interval, 3.12-38.59; P < .001). Having a higher (better) VF score was associated with a lower likelihood of accepting refractive services (odds ratio, 0.30; 95% confidence interval, 0.10-0.91; P = .03) (Table 5). Age, sex, presenting/best-corrected visual acuity (analyzed separately and together), hyperopia vs myopia, astigmatism, and reading (ever vs never) were not significantly associated with uptake of refractive services.
Among 109 examined subjects who had not yet undergone bilateral cataract surgery, 80 (73.4%) were thought to have visually significant cataract in the eye not operated on. These persons included 71% (71 of 100) of subjects with low visual acuity (presenting visual acuity ≤ 6/18) and 80% (48 of 60) of patients with blindness (presenting visual acuity ≤ 6/60) in either eye. Of the 80 subjects with cataract in the fellow eye, 18 (22.5%) accepted second-eye surgery, 31 (39%) refused, and the remainder were undecided. Reasons for refusing or being undecided about surgery generally focused on lack of resources: 59% of refusing subjects indicated that they had insufficient money or that transport was a problem. Fewer subjects indicated that they felt “no need” (27%) or that they were “too old” (19%). A total of 40 subjects had already undergone bilateral cataract surgery at Sanrao at the time of examination. Comparing the total of 58 patients accepting or having already undergone second-eye surgery at Sanrao with the 31 refusals, younger patients were more likely to accept surgery (P = .004), but sex, presenting visual acuity in the first eye operated on, reading (ever vs never), and VF score were not associated with uptake of second-eye surgery in the multivariate model (Table 6).
Self-reported VF was extremely high in this cohort of subjects having undergone cataract surgery 10 to 14 months previously. This is in marked contrast to poor VF reported in 2 other studies of cataract surgical outcomes in rural China.1,2 In fact, poor visual outcomes, including low VF and quality of life, have been reported after cataract surgery throughout rural Asia.1- 6,11,12 Results in this study are comparable with the best VF outcomes reported at major urban centers such as Aravind Eye Hospital.9
The high levels of VF in the current study were largely driven by excellent postoperative presenting visual acuity. As previously reported,14 83% of subjects had presenting visual acuity of 6/18 or better, results comparable with figures for postoperative visual acuity from large studies in Sweden,17 the United States,18 and the United Kingdom.19 Bilateral (as opposed to unilateral) surgery was also associated with better VF outcomes in the current study. The additional visual and functional benefits of second-eye cataract surgery are well established,20,21 including in Chinese populations.22 The possibility that patients with better visual results in the first eye were more likely to undergo surgery in the second eye cannot be excluded.
A principal goal of the current study was to understand the utilization and perceived need for refractive services in this population, as well as factors determining the willingness to accept spectacle prescriptions. These parameters are of importance for a sustainable cataract surgical program such as Project Vision/Caring is Hip for 2 reasons: postoperative refraction has been identified as a critical factor in optimizing visual outcomes after cataract surgery in rural areas2- 5,11,12 and the sale of spectacles is an important potential source of revenue.
Refraction could potentially reduce the number of eyes operated on with postoperative presenting distance acuity of 6/18 or better by 74% (26 of 35) in the SCOUTS cohort. Nearly 87% of patients in the cohort could improve significantly at near or distance in at least 1 eye with refraction. However, the uptake of refractive services was low, with only about a third of patients whose near or distance acuity improved in either eye by 2 or more lines being willing to accept a written prescription. Lack of a perceived need, rather than cost, was the principal barrier.
However, a significant proportion of subjects in this study (some 40%) already owned spectacles. Those with spectacles already were 11 times more likely to be willing to accept a prescription for new glasses. This suggests that there exists in rural China a modest proportion of persons with a strong interest in refractive services. Correction of near vision is of particular importance to this group: more than half of those with glasses, roughly a quarter of all subjects in the study, had reading glasses only. Persons with reading glasses were also more likely to use their spectacles regularly than those with distance or other types of glasses.
Recent research has demonstrated the importance of presbyopia in the developing world. The Andhra Pradesh Eye Disease Study reported a prevalence of more than 50% for persons older than 30 years in South India, with rural residents at a 50% increased risk.23 Studies in Tanzania have identified a prevalence of more than 60% among persons 40 years and older24; presbyopia was associated with significantly increased difficulty with daily tasks.13 The problem of presbyopia in rural China will only increase because the number of persons 50 years and older is predicted to triple nationwide between 1995 and 2050.25 Vision programs hoping to achieve cost recovery through the sale of spectacles to adults in rural China will need to focus on reading glasses, though the relatively low proportion of uptake and the prevailing low price for readers will pose important limitations.
Finally, in assessing the relative importance of near vision in this rural Asian cohort, self-reported VF was strongly associated with presenting distance vision in the eye that was operated on, but unassociated with presenting near vision. It is possible that this may have been driven to some extent by the composition of the VF test we used16: 3 items related explicitly to distance vision, 1 to near, and 2 to intermediate vision. Remaining items related to color vision, glare, transitions between light and dark, etc.
Though bilateral cataract surgery was common in this cohort, lack of cataract surgery remained an important cause of unilateral blindness and low vision, accounting for 80% of persons blind in 1 or more eyes and 71% of those with unilateral or bilateral low vision. In addition to 67 patients (40 operated on bilaterally at Sanrao, 27 in whom one eye was operated on at Sanrao and one was operated on elsewhere) who had already undergone bilateral cataract surgery, 18 subjects with visually significant cataract in the fellow eye expressed a willingness to undergo second-eye surgery. Among 176 examined subjects, the rate of bilateral pseudophakia could thus be as high as 48%. Younger patients were significantly more likely to be willing to undergo second-eye surgery.
Patients having undergone bilateral surgery had better VF in this cohort, as mentioned earlier. The willingness of nearly one-half of these rural-dwelling Chinese persons to pay for surgery in both eyes has important implications not only for optimal reduction in blindness and low vision, but also for the sustainability of programs such as this one, which are largely dependent on surgical revenue. As compared with the situation with uptake of refractive services, the principle barrier to second-eye surgery in this cohort is financial. Tiered pricing, allowing for an increased number of surgeries to be offered at reduced prices, may be a solution.
The limitations of SCOUTS must be understood in assessing the meaning of these results. First, prospective data on preoperative VF do not exist for the SCOUTS cohort, making it impossible to document with certainty the impact of surgery on VF. Second, the proportion of eligible patients who could be contacted, and then of those contacted who could be examined, were both about 75%. Approximately 75% of the overall sample underwent in-person or telephone interviews. The possibility must thus be acknowledged that interviewed and examined patients may not be representative of patients undergoing surgery at Sanrao in general. The likelihood of meaningful bias is somewhat reduced, however, by the fact that examined and interviewed patients did not differ significantly from those who could not be contacted with regard to age, sex, preoperative presenting visual acuity, or day 1 postoperative presenting visual acuity in the eye operated on.
SCOUTS provides unique information on the visual acuity, VF, spectacle use, and uptake of postoperative services among a clinic-based sample of patients undergoing cataract surgery in rural China. The importance of these data lies in the high relevance of the Caring is Hip Sanrao model to cataract blindness in China, which has among the lowest cataract surgical rates in Asia at 446 cases per million population per year in 2004.26 This represents only 24 cases annually for each of China's 23 000 ophthalmologists.27 Few of these ophthalmologists currently practice in the countryside, where 60% of China's population resides. The Sanrao model is based on providing high-quality surgery at prices that are affordable to the local population but sufficient to sustain programs at rural government hospitals. This offers a unique solution to the twin problems of underfunded rural facilities and unavailability of surgical services in the countryside. Data on the willingness of patients to take up the full array of sight-improving services offered by such rural facilities will be critical to program planners hoping to replicate this important model.
Correspondence: Nathan G. Congdon, MD, MPH, Department of Ophthalmology and Visual Sciences, The Chinese University of Hong Kong, Hong Kong Eye Hospital 3/F, 147K Argyle St, Kowloon, Hong Kong, China (email@example.com).
Submitted for Publication: May 1, 2007; final revision received September 8, 2007; accepted September 9, 2007.
Published Online: October 22, 2007 (doi:10.1001/archophthalmol.125.11.eeb70014).
Author Contributions: Drs Congdon and Lam contributed equally to the study.
Financial Disclosure: None reported.
Funding/Support: Funding for this study and for Project Vision (the eye care component of the Caring is Hip medical relief program) was provided by the Li Ka Shing Foundation.
Additional Contributions: Frieda Law, MD, assisted in the execution of this project.