Menu

  1. Home
  2. /
  3. Col.Rangachari Session, Free Papers, Optics / Refraction /..., Best Papers
  4. /

Share this post

RANGACHARI SESSION VIDEO

 

Authors:

Parikshit Gogate, FRCS (Edin)6,  7, 8, Priya AdhiseshaReddy, MBA1; Prof Nathan Congdon, MD2, 3, 4; Graeme MacKenzie, PhD5; Qing Wen, PhD2; Catherine Jan, OD9; Prof Mike Clarke, DPhil2; Jordan Kassalow, OD10; Ella Gudwin, MA10; Prof Ciaran O’Neill, PhD2; Ling Jin, MS4; JianjunTang, PhD2; Prof Ken Bassett, MD11, 12; David H. Cherwek, MD3; Rahul Ali, MBBS MPH3

Affiliations:

1Aravind Eye Hospital and Post-graduate Institute of Ophthalmology, Pondicherry, India. 2Queen’s University Belfast, Belfast, United Kingdom. 3Orbis International, New York, NY, United States. 4Zhongshan Ophthalmic Center, Guangzhou, China. 5Clearly, London, United Kingdom. 6Community Eye Care Foundation, Pune, India. 7Padmashri D.Y.Patil Medical College, Pimpri, Pune, India. 8Lions NAB Eye Hospital, Miraj, India. 9School of Psychological and Cognitive Sciences, Peking University, Beijing, China. 10VisionSpring, New York, NY, United States. 11University of British Columbia, Vancouver, BC, Canada. 12Seva Canada, Vancouver, BC, Canada.

Abstract

Background:Presbyopia, age-related decline in near vision, is the commonest cause of vision impairment globally, but no trials have assessed workplace effects. We studied impact of glasses on productivity of presbyopic tea-workers.

Methods:Tea-pickers aged≥40 years in Assam, India, withunaided near visual acuity (NVA)<6/12 in both eyes, correctable to 6/7·5 with near glasses; unaided distance vision ≥6/7·5; and no eye diseasewere randomised (1:1, stratified by age, sex,productivity) toreceive free glassesoptimizing NVA at working distance (cost including delivery: US$10.20/person), either immediately (Intervention) or at closeout (Control). Main study outcome (investigator-masked, intention to treat) was difference between groups in change in mean daily weight of tea picked (productivity), between 4-week baseline and 11-week evaluation period. Glassescompliance was assessed at 7 un-announced visits.

Results:Among 2,699 permanent workers, 1,297 (48·1%) met age criteria and consented for eye examinations(3-15 July 2017), and 751 (57·9%) met vision criteria and were randomized to Intervention (n=376, 50·1%, mean age 47·2 [range 40-59] years, 77·9% women) or Control (n=375, 49·9%, mean age 47·1 [40-61] years, 78·1%women). Groups did not differ substantially in baseline characteristics. No participants owned glasses at baseline, 94·1% (n=707) received allocated intervention, all were followed up and analyzed. Mean Intervention group productivity increased from 25·0 to 34·8 kg/day (9·84 kg/day), significantly more than Control (26·0 to 30·6, 4·59 kg/day), a difference of 5·25 kg/day [21·7%, Effect size 1·01] (95% Confidence Interval [CI] 4·50, 5·99, P<0·001).Intervention group compliance with study glasses was 70·9%.Regression model predictors of greater productivity increase included Intervention group membership (5·25 kg/day, 95% CI 4·60, 5·91, P<0·001) and, among Intervention participants, older age (P=0·039) and better compliance (P<0·001).

Conclusion:A substantial productivity increase was achieved in this rural cohort with little costand high intervention uptake.

Introduction

The Sustainable Development Goals (SDGs),1central tothe United Nations’ global development agenda, call for an end to poverty (SDG #1) and the promotion of health for all (SDG #3). It is a basic tenet of global health policy that good health and productive work are causally linked,2 and yet the effects on labor productivity of very few health interventions have been evaluated inrandomized trials in low and middle-income countries (LMICs).3–8 Most such trials involve dietary supplementation to improve nutritional status.3,5–8These interventions have generally shown uncertain or non-significanteffect sizes,3,5–8 and require sustainable distribution pathways and long-term adherence with daily dosing,both challenging in low-resource areas.

Low-cost, sustainable and effective health interventions are needed to increase work productivity and reduce poverty in LMICs. With the global population aging rapidly, and labor participation rates in LMICs declining above age 45,9 health strategies which support productive employment (SDG #8)1 among older workersare of particular interest. Gender equality (SDG #5)1 is also highly-relevant to poverty alleviation, as increasing workforce participation and productivity among women results in faster economic growth.10

Presbyopia is the essentially universal condition in which un-aided near vision declines with aging. Presbyopia usually becomes functionally apparent around age 40 years, and is essentially complete by 55,11meaning that presbyopia is highly-prevalent during the working years.12 The number of persons affected by presbyopia globally exceeds 1 billion, making this the world’s most common cause of vision impairment.13Although presbyopia is safely, effectively and inexpensivelytreated with glasses, rates of optical correction in LMICs are as low as 10%.14,15The global productivity loss due to un-corrected presbyopia has been estimated at US$25 billion,16 and presbyopia is associated with significant impairment in activities of daily living.17 However, there are no published trials examining whether optical correction of presbyopia with glasses improves work productivity.

We carried out a randomized, investigator-masked trial to determine whether provision of free glasses to presbyopic, mostly-female tea workers in India aged ≥40 years improved work productivity. Our hypothesis was that membership in the Intervention group would lead to significantly greater increases in productivity compared to Controls, measured as the group difference in change in mean daily weight of tea picked between a baseline and subsequent evaluation period.

Methods

Study design

The PROSPER (PROductivity Study of Presbyopia Elimination in Rural-dwellers) investigator-masked, randomised trial was carried out on three tea plantations in Assam, India. The protocol was approved by Ethics Committees at Lions National Association of the Blind (NAB)Eye Hospital (Miraj, India) and Queen’s University Belfast (Belfast, UK; approval for data analysis). The tenets of the Declaration of Helsinki were followed throughout.

Participants and setting

Unlike the rest of the year, the amount of tea picked during the high season in Assam, India (June–October)is generally limited by the worker, rather than by the rate of plant growth, and income is thus tied to daily productivity as an incentive during this period. All permanent workers aged ≥40 years as of 31 December 2017at three tea gardens in Assam (Kellyden, Nonoi, Sagmootea)owned by Amalgamated Plantation Private Ltd (APPL), andwho had worked for APPL for at least one year and for ≥10 days in the previous month, were invited to undergo a free eye examination (see below).Study personnel obtained oral informed consent from all workers undergoing the eye examination. The study inclusion criterion based on this examinationwas the presence in both eyes of presbyopia, defined as habitual Near Visual Acuity (NVA)≤6/12at 40cm, correctable with spherical (non-astigmatic) near glasses to ≥6/7·5. (Astigmatic eyes have a non-spherical shape, and require glasses which are more expensive and time-consuming to make).Exclusion criteria were:

  • Current ownership of near glasses capable of improving NVA to≥6/12 in either eye.
  • Presence in either eye of refractive error, other than presbyopia, associated with unaided distance vision<6/7.5.
  • Visually-significant eye disease (e.g.: cataract) in either eye detected during the eye examination.
  • Low likelihood of completing follow-up in the study.

Randomization and masking

Consenting participants eligible for the trial were divided into eightstrata according to age (<50 years, ≥50 years), sex and baseline work productivity during June 2016 (<median, ≥median). Participants in each stratum were randomized 1:1with block size of six to one of the following groups:

Intervention group: Immediately receivedfree, spherical presbyopic glasses providing the best NVA in each eye at the participant’s usual working distance.

Control group: Received similar glasses after the 11-week evaluation period.

Allocation concealment was achieved by having the randomization sequence generated by the study statistician, who had no contact with participants or study personnel, at the Clinical Trials Unit of the Zhongshan Ophthalmic Center (Guangzhou, China)using an online random number generator (www.randomization.com), and concealing it until a worker was determined eligible and agreed to participate. The field team consisted of 4experienced optometrists, 3 nurses, 2 data assistants and a screening coordinator, all employed byVisionSpring, an eye health non-governmental organization (NGO). This team had a list provided by APPL of potential participants and their age and productivity in June 2016, and carried out eye examinations and distributed glasses to eligible participants in the Intervention group from 3-15 July 2017. Study personnelaccessed the random assignment for each participant according to the correct age-sex-productivity stratum only at the time of enrolment, to facilitate concealment of allocation.

Study personnel carrying out the eye examination and facilitating randomisation and distribution of glasses had no further contact with participants until glasses were distributed to the Control group after the end of the trial.It was not practical to mask participants to their study group assignment, as the investigators did not feel that providing sham glassesto the Control group was ethical. However, APPL staff measuring the weight of tea picked (main study outcome) were masked to workers’ group assignments. Intervention participants removed their glasses before proceeding to the weighing station. One employee received the sack of tea and suspended it on an electrical scale (Easyweigh, Applied Data Logix, Chennai, India), while a second, also masked to workers’ group assignments, swiped the participant’s work identification card. The scale automatically measured and recorded the weight in the APPL database, without staff input.

Procedures

Vision assessment

Consenting workers meeting age and work criteria outlined above, first underwent measurement of distance visual acuity, in each eye separately,by a study nurse using a log of the minimum angle of resolution (logMAR) chart in a well-lit area. Study optometrists next measured unaidedNVA (without near glasses)in each eye separately using a tumbling E chart based on the “reduced” Early Treatment Diabetic Retinopathy Study (ETDRS) configuration,18held at a distance of 40cm at the level of the participant’s chin.

Refraction and distribution of glasses (Intervention)

Workers meeting above criteria for near and distance vision underwent measurement of glasses power (refraction) by an optometrist at theusual working distancein each eye. They were directed to stand in front of a tea bush selected for measurement purposes, and adopt their usual stance for picking tea, covering the left and then the right eye with the right hand.Working distance (distance from each eye to the top of the bush) was measured, and near refraction was carried out with a trial frame, the end point being the highest plus sphere power in each eye enabling the participant to identify 2-3 leaves and a bud appropriate for picking. Best-corrected NVA was then assessed in each eye separately, using the glasses power obtained at the working distance,with the above-described near chart at 40cm. This measurementwas not assessed at the working distance because near vision charts are designed for useat 40cm, and results are inaccurate at other distances. Glasses were provided to the Intervention group immediately after the eye examination, with directions to begin wearing them on 24 July 2017, when the trial began, participants were formally enrolled, and collection of data in the evaluation period commenced.

Questionnaires

The study team orally administered questionnaires to all workers taking part in the screening eye examination. Data were collected on potential determinants of productivity and compliance withglasses, including age, sex, marital status, baseline glasses ownership, years working as a tea picker, and work attitudes (importance of maximizing income by picking as much tea as possible, whether or not tea picking was the main source of family income). Participants’ height was also measured as a potential determinant of productivity.

Service delivery

Workers with unaided distance visual acuity<6/18 (not eligible for the study) underwent refraction, and later received free distance glasses if visual acuity improved to ≥6/12. Near glasses were also later provided to workers requiring astigmatic correction. Those whose distance visual acuity did not improve with glasses underwent examination by an optometrist, using a flashlight and direct ophthalmoscope, without dilation of the pupil. Workers with visually-significant cataract were referred for free surgery through the local District Blindness Control Society, and those with other conditions were referred to a local eye hospital (ERC Eye Care, Jorhat, Assam), with medical expenses assumed by APPL.

Assessment of outcomes and compliance

Data for daily weight of tea picked by participants in both groups was assessed by masked APPL employees as described aboveduring an 11-week evaluation period from 24 July to 7 October 2017.Baseline productivity datawas available retrospectively for participants in both groups for June 2017 (4 weeks). (Randomization was stratified on June 2016 productivity data, as June 2017 data were not yet available). The main study outcome was the difference between randomized groups in change in mean daily weight of tea picked between baseline and evaluation periods. Secondary outcomes were visual quality of life, assessed using the National Eye Institute Visual Function Questionnaire-25 (VFQ-25)19on a single occasion for both groups during the evaluation period, 2-8 August 2017);observed wear of glasses in the Intervention group and purchase/wear of glasses in the Control group. Glasses wear in both groups was assessed by study personnel during 7un-announced visits, carried out weekly during the evaluation period except in the event of holidays or flooding, and recorded as Glasses Worn, Glasses not Worn or Worker not Present. APPL field supervisors encouraged but did not require Intervention group participants to wear study glasses. Daily work attendance for both groups was assessed during the evaluation period based on APPL records. Subjective usefulness of the study glasses, willingness to recommend them to other workers and pay for them if lost/broken (5-point Likert scale) and amount willing to pay wereassessed for the Intervention group, but were not pre-specified study outcomes.

Statistical analysis

With two-sided significance level of P=0·05, power of 80%, a mean of 25·0 kg/day(standard deviation: 5·0) of tea picked during the baseline month (based on APPL records), and allowing for 20% loss to follow-up, a sample size of 160 persons (80 per group) would be sufficient to detect a 10% greater increase in daily mean weight of tea picked in the Intervention group compared to Controls (main outcome). In order toconduct adequately-powered age-stratified analyses,we sought to recruit 700 participants.

  • Give details of main comparative analyses
  • State which participants were included in primary and compliance analyses (e.g., by intention to treat…)
  • State statistics program and version number
  • List trial registration number and registry

Role of funding source

The funder of the studyhad no role in study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.

Results

Among 2,699 permanent tea pickers at the three plantations, 1,301 (48·2%) met age and work criteria and were eligible for the screening eye examination (Figure 1). Among these, 1,297 (92·8%) completed the examination, while 101 (7·22%) were absent (n=63, 4·51%) or declined (n=38, 2·72%). Based on the examination, 751 (57·9%) workers were eligible for the trial and underwent stratification and randomization, while 546 (42·1%) were excluded (causes detailed in Figure 1).In total, 376(50·1%) participants were allocated to Intervention, and375 (49·9%) to Control. Among Intervention and Control participants, 361 (96·0%) and 346 (92·3%) respectively received their allocated treatment, while all completed follow-up and were analyzed in intention-to-treat fashion. (Figure 1)

Participants in both groups were primarily (>75%) women, with a mean age of 47 years (Table 1). No participants had glasses at baseline, and most had modest presbyopia, with <20% in both groups having NVA<6/18. Nearly all participants in both groups had normal NVA (6/6 at 40cm) with study glasses (which were given to Controls only after the evaluation period). In both groups, mean time working as a tea picker was >25 years, and all who answeredagreed or strongly agreed that “It is important to maximize my income by picking as much tea as possible,” and “Picking tea is the main source of family income during the high season.” Two participants (0·027%) had missing data for starred variables listed in Table 1. Baseline mean daily weight of tea picked (June 2017) differed between Control (26.0 kg/day) and Intervention participants (25.0 kg/day, P<0.001), as stratification on the most recent available data (June 2016) did not fully balance the groups.

Daily weight of tea picked (main study outcome) in the Intervention group increased from 25·0 to 34·8 kg/day between the baseline and the evaluation periods. This increase of 9·84 kg/day (95% Confidence Interval [CI] 9·27, 10·4) was greater than that for Controls (26·0 to 30·6 kg/day, increase: 4·59, 95% CI 4·10, 5·07). The between-group difference (5·25 kg/day, 95% CI 4·50, 5·99, P<0·001) was statistically significant, and equivalent to a 21·7%relative productivity increase, with an effect size (Cohen’s d) of 1.01. (Table 2). In multiple regression models, variables significantly associated with greater productivity increase included Intervention group membership (5·25 kg/day, 95% CI 4·60, 5·91, P<0·001), female sex (4·61 kg/day, 95% CI 3·65, 5·57, P<0·001) and work attendance during the evaluation period (0·367 kg/day for 10% increased attendance, 95% CI 0·098, 0·635P=0·008). (Table 3) Years working as a tea picker (P=0·042) andIntervention groupglasses compliance (P<0·001) were significantly associated with greaterproductivity increase in univariate analyses, but were excluded from multivariate models due to collinearity withage and group assignment respectively.

Presbyopia is strongly age-related, and the proportion of all participants with unaided NVA<6/12 in the better-seeing eye rose significantly with age: 51·6% (146/283) at40-44 years, 65·6% (128/195) at 45-49 years, 75·8% (207/273) at 50+ years (P<0·001). (Similar trends were present in both studygroups, data not shown). In multivariate models, the association between age and productivity increase behaved differently between Intervention and Control groups: increasing age was associated with significantly greater productivity increase (0·112 kg/day per year of age, P=0·039) in the Intervention group, but negatively associated with productivity increase (-0·158 kg/day per year of age, P<0·001) among Controls. Thus, the difference between groups in change in productivity rose with increasing age: 40-44 years (4·24 kg/day [16·4% of baseline], 95% CI 3·19, 5·28); 45-49 years (6·00 kg/day [23·2%], 95% CI 4·78, 7·25), 50+ years (7·96 kg/day [32·1%], 95% CI 6·63, 9·92, P<0·05).

No Control group participants had purchased glasses during compliance assessments.

Discussion

We observed a significantly greater increase in productivity of >20%among Intervention group members in this largely-female, rural cohort, using a low-cost (US$10) intervention which was highly-acceptable to participants. The effect size of 1·01 fell between “large” (>0·8) and “very large” (>1·2).20

Presbyopia is strongly associated with aging,11and we observed a significant interaction between age and study group for the main study outcome (Table 3).Older participants had significantly greater productivity increase in the Intervention group, while older age was associated with smaller productivity rises among Controls. Larger proportional gains among older Intervention participants were apparently due to their lower baseline productivity (40-44 years: 25·6 kg/day [95% CI CC, DD]; 50+ years 23·9 kg/day [95% CI BB, CC, P=0·021]), potentiallydue to uncorrected presbyopia. Their deficit compared to younger workers disappeared during the evaluation period (40-44 years: 35·4 kg/day [95% CI CC, DD]; 50+ years 35·3 kg/day [95% CI FF, GG, P=0·80]), presumably due to the effects of corrective eyewear.

Regarding Controls, we hypothesize that observed increases in productivitywere due to more rapid growth of tea at the peak of the high season during the evaluation period (July-October), compared to the baseline period in June. Older, more presbyopic workers were apparentlyless able than their younger colleagues to take advantage of higher tea yields to improve their daily harvest, resulting in lower productivity increases with age among Controls. This strong interaction of age with study groupadds to the biological plausibility of our results.

We carried out two systematic reviews of studies performed in LMICs: first,on the impact of presbyopia and second on health interventions to improve work productivity. The former searched the MEDLINE, Embase and WHO Library databasesfrom 19-26 November 2017for articles in any language appearing after 1 January 1975, using the following combinations of terms together with a list of LMICs: “near vision” AND “impairment”; “prevalence” AND “presbyopia”; “presbyopia” OR “near vision” AND “correction”; “presbyopia” AND “correction” AND “productivity”; “presbyopia” AND “correction” AND “quality of life” AND “activity”. Only publications reporting on actual (not modelled) impacts of presbyopia were included. Population studies of cohorts aged >40 years in rural China21 and Tanzania17 reported two and eight-fold increases respectively in difficulty with near activities of daily living among persons with presbyopia, though neither study assessed impact of correction with glasses (low-quality data). Among 187mostly-presbyopic persons aged ≥40 years in Zanzibar, effect size was substantial for variouswork-related activities of daily living (1·6-3·8, all P<0·001) (moderate-quality data).22We identified no randomised trialsin this review.

In our second review, we searched MEDLINE, Embase and Econlit databases on 16 November 2017 for studies published in any language without date restrictions, using53 terms (Annex 1) designed to identify 2 types of investigations conducted in LMICs:randomised trials testing impact of health interventions on labour productivity, and studies (including non-trials) of the effect of vision care on productivity.

An iron supplementation trial among 199 anaemic tea workers in Sri Lanka3 reported a significant effect on the main trial outcome (daily weight of tea picked) for the Intervention group only, but found no between-group difference. No significant difference between any groups in labour productivity (weight of tea picked)was reported in a factorial-design trial of iron and anthelmintics among 553 Bangladeshi tea workers.8 An Indonesian trial of iron supplementation involving 302 rubber plantation workers7reported a significant (4·32 kg, 14·5%, P<0·05) difference between study groups in productivity for tappers, but not weeders. Among 80 Chinese mill workers receiving iron or placebo for 12 weeks,5 a significant (5·2%, P<0·05) increase in productivity was observed in the Intervention group, but there was no between-group productivity difference. A Kenyan trial of two levels of caloric supplementation among 224 road construction workers6 reported a non-significant productivity gain in the group receivinggreater supplementation, but no between-group comparison was reported. Provision of mosquito netting in a cluster-randomised trial involving 516 farmers in a malarial area of Zambia4 was associated with an improved harvest value of US$76 (14·7%, P<0·05) compared to Controls, though Intervention group farms were larger and more productive at baseline. In summary, few recent trials (one since 2005) have assessed impact of health interventions on productivity in LMICs. Existing studies have many methodologic flaws, and only two have significant effect sizes, providing low-moderate quality evidence for a modest effect of iron supplementation, and possibly mosquito netting, on work productivity.

Our review identified no trials of vision care interventions with productivity outcomesin LMICs. However, two prospective, non-randomised studies reported significant productivity improvements inparticipants accepting cataract surgery (moderate quality). Among 294 Indian participants, a significant (from 44·7 to 77·7%, P<0·001) increase was observed in income-generating behavior,23 while a controlled, multi-countrystudy24 reported significant (P<0·001) increase in productive activities amongoperated participants in all settings compared to Controls. Effects observed in the latter study largely persisted at 6 years.25A recent review on productivity and neglected tropical diseases26(moderate-quality) concluded: “The largest impact on productivity loss…seems to be due to blindness from onchocerciasis and severe schistosomiasis.”

Evidence that eye care, particularly cataract surgery, can have a substantial impact on economic well-being is consistent with our results. The relative productivity increase in our Intervention group, approximately 20%, was as large or larger than effect sizes reported for other LMIC health intervention trials.3–8Compared to correcting presbyopia, equipment, consumables and training required for cataract surgery are far greater, and achieving optimal visual results in LMIC settingsis more difficult.27Presbyopia also has a higher prevalence13 and earlier onset than age-related cataract, becoming apparent during prime working years,12 when economic impact is greatest.

Other aspects of our findings also suggest that glasses delivery to older workersmay be an appealing strategy for poverty alleviation. Our intervention was inexpensive, and could potentially be sustained by employers benefitting from productivity increases, or by workers themselves. Over 90% of Intervention participants would pay for glasses, consistent with reports from Zanzibar that respondents would pay 34% of monthly income for presbyopic glasses.22Achieving good adherence with presbyopic glasses may be less challenging than long-term nutritional supplementation, the best-studied health intervention for LMIC productivity improvement,3,5–8which requires un-interrupted distribution and regular compliance. Our observed wear rates of study glasses exceeded 80% by closeout, consistent with reports of 96% acceptance22 and 94% one-year retention28 of presbyopic glasses in LMICs. Longer-term follow-up of working cohorts is needed to better characterise glasses compliance.

Treatment of presbyopia to improve labour productivity among older workers is appealing in view of aging of the global population.Age-related decline in unaided near vision is essentially universal, so the pool of potential beneficiaries is large. Our data suggest correction of even modest presbyopia, present among>50% of workers aged ≥40 years in our study, is associated with significant productivity gains. Correction of presbyopia might also promote workforce participation among older workers. A study of Kenyan tea workers medically retired due to Acquired Immune Deficiency Syndrome (AIDS)29 reported decreased productivity of 16-18% in the last 18 months of work. Productivity increasesof the magnitude observed in the current trial might thus be sufficient to prevent presbyopic labourers from leaving the workforce.

Inferences about applicability of our results to other work settings are limited by incomplete data. It is unclear what proportion of economically-productive activity among older persons in LMICs is sufficiently vision-dependent that correction of presbyopia would be useful. Existing reports21–22from LMICs suggest uncorrected presbyopia affects many activities relevant to economic behaviour, including reading, writing, cooking, mobile phone use, sewing, weeding, recognisingmoney and utilisingtools. Additional trials are needed in other work settingsto broaden the evidence base for glasses delivery programs.

Other limitations include the fact that our study did not assess glasses safety. Study glasses were made of polycarbonate, widely used for workplace eye protection.So it is unlikely they posed an important work hazard. For practical and ethical reasons, participants were not masked to study group assignment, leaving open the possibility of placebo effects. In baseline questionnaires, 100% of participants indicated their desire to maximise earnings through greater productivity and their dependence on tea picking as a main income source. It thus seems likely workers were already making a maximal effort at harvesting tea, and were therefore less prone to large placebo effects. The significant and biologically-plausible interactionbetween age and study groupis also difficult to explain as a placebo effect, though thiscannot be definitely excluded. Due to field team errors, 44 (5·85%) participants did not receive their allocated treatment. Intention-to-treat analysis reduced observed between-group differences in the main outcome, but preserved benefits of randomization. A modest but statistically-significant baseline difference in between-group productivity resulted from our stratifying on available 2016 rather than 2017 data, though the difference-of-difference analysis reduced any impact on results.

Study strengths, besides the randomised, controlled design, included completeness of data for the main study outcome and its determinants, no losses to follow-up and selection of a rural, low-incomepopulation highly relevant to assessing poverty alleviation strategies in LMICs.

Acknowledgments: Funding for the study was provided by Clearly World (https://clearly.world/). Prof Congdon is supported by the Ulverscroft Foundation (UK). The authors acknowledge the support and cooperation of Conrad Dennis and colleagues at the Amalgamated Plantation Private Ltd (APPL), without whom the study would have been impossible·

Figure legends

Figure 1: Enrollment and progress of participants through the study (CONSORT flowchart)

Figure 2: Observed compliance with glasses wear (Intervention group only) and percentage gain in productivity in the Intervention and Control groups during the evaluation period, stratified by age. (Error bars indicate plus or minus two standard deviations).

References.

  1. United Nations Sustainable Development Knowledge Platform. The Sustainable Development Goals. https://sustainableun.org/sdgs. Accessed 5 December 2017.
  2. World Health Organization. Programmes. Health and Development. https://www.who.int/hdp/en/.Accessed 5 December 2017.
  3. Edgerton VR, Gardner GW, Ohira Y, Gunawardena KA, Senewiratne B. Iron-deficiency anemia and its effects on worker productivity and activity patterns. BMJ1979; 2: 1546–49.
  4. Fink G, Masiye F. Health and agricultural productivity: Evidence from Zambia. J Health Econ2015; 42: 151–64.
  5. Li R, Chen X, Yan H, Deurenberg P, Garby L, Hautvast JGAJ. Functional consequences of iron supplementation in iron-deficient female cotton mill workers in Beijing, China. Am J ClinNutr1994; 59: 908–13.
  6. Wolgemuth JC, Latham MC, Hall A, Chesker A, Crompton DWT. Worker productivity and the nutritional status of Kenyan road construction laborers. Am J ClinNutr1982; 36: 68–78.
  7. Basta SS, Soekirman, Karyadi D, Scrimshaw NS. Iron deficiency anemia and the productivity of adult males in Indonesia. Am J ClinNutr 1979; 32: 916–25.
  8. Gilgen D, Mascie-Taylor CGN, Rosetta L. Intestinal helminth infections, anaemia and labour productivity of female tea pluckers in Bangladesh. Trop Med Int Health 2001; 6: 449–57.
  9. International Labor Organization. Statistics and Databases. Key Indicators of the Labor Market (KILM). https://www.ilo.org/ilostat/faces/oracle/webcenter/portalapp/pagehierarchy/Page3.jspx?MBI_ID=142&_adf.ctrl-state=ny099njh7_62&_afrLoop=14921558630049&_afrWindowMode=0&_afrWindowId=null#!%40%40%3F_afrWindowId%3Dnull%26_afrLoop%3D14921558630049%26MBI_ID%3D142%26_afrWindowMode%3D0%26_adf.ctrl-state%3Djj1iv5fxz_21.Accessed 5 December 2017.
  10. Organization for Economic Cooperation and Development (OECD), Gender Equality in Education, Employment and Entrepreneurship: Final Report to the MCM 2012. http: //www.oecd.org/employment/50423364.pdf. p. 17. Accessed 5 December 2017.
  11. Glasser A, Kaufman PL. Accommodation and presbyopia. In: Kaufman PL, Alm A, eds. Adler’s Physiology of the Eye. St Louis: Mosby; 2002: 197–233.
  12. He M, Abdou A, Ellwein LBet al.Age-related prevalence and met need for correctable and uncorrectable near vision impairment in a multi-country study. Ophthalmology 2014; 121: 417–22.
  13. Bourne RR, Flaxman SR, Braithwaite Tet al., on behalf of the Vision Loss Expert Group.Magnitude, temporal trends, and projections of the global prevalence of blindness and distance and near vision impairment: a systematic review and meta-analysis. Lancet Glob Health 2017; 5: e888–97.
  14. He M, Abdou A, Naidoo KSet al.Prevalence and correction of near vision impairment at seven sites in China, India, Nepal, Niger, South Africa, and the United States. Am J Ophthalmol2012; 154: 107–16.
  15. Goertz AD, Stewart WC, Burns WR, Stewart JA, Nelson LA. Review of the impact of presbyopia on quality of life in the developing and developed world. ActaOphthalmol 2014; 92: 497–500.
  16. Frick KD, Joy SM, Wilson DA, Naidoo KS, Holden BA. The Global Burden of Potential Productivity Loss from Uncorrected Presbyopia. Ophthalmology 2015; 122: 1706–10.
  17. Patel I, Munoz B, Burke AGet al. Impact of presbyopia on quality of life in a rural African setting.Ophthalmology 2006; 113: 728–34.
  18. Rosser DA, Laidlaw DA, Murdoch IE. The development of a “reduced logMAR” visual acuity chart for use in routine clinical practice. Br J Ophthalmol2001; 85: 432–6.
  19. Mangione CM, Lee PP, Gutierrez PR, Spritzer K, Berry S, Hays RD; National Eye Institute Visual Function Test Questionnaire Field Test Investigators. Development of the 25-item National Eye Institute Visual Function Questionnaire. Arch Ophthalmol2001; 119: 1050–8.
  20. Sawilowsky S. New effect size rules of thumb. J Modern Applied Statistical Methods2009; 8:467–74.
  21. Lu Q, Congdon N, He X, Murthy GV, Yang A, He W. Quality of life and near vision impairment due to functional presbyopia among rural Chinese adults. Invest Ophthalmol Vis Sci 2011; 52: 4118–23.
  22. Laviers HR, Omar F, Jecha H, Kassim G, Gilbert C. Presbyopic spectacle coverage, willingness to pay for near correction, and the impact of correcting uncorrected presbyopia in adults in Zanzibar, East Africa. Invest Ophthalmol Vis Sci 2010; 51: 1234–41.
  23. Finger RP, Kupitz DG, Fenwick E, et al.The Impact of successful cataract surgery on quality of life, household income and social status in south India. PLoS ONE2012; 7: e44268.
  24. Polack S, Eusebio C, Mathenge W, et al.The impact of cataract surgery on activities and time-use: Results from a longitudinal study in Kenya, Bangladesh and the Philippines. PLoS ONE2010; 5: e10913.
  25. Danquah L, Kuper H, Eusebio C, et al.The long termimpact of cataract surgery on quality of life, activities and poverty: Results from a six year longitudinal study in Bangladesh and the Philippines. PLoS ONE2014; 9: e94140.
  26. Lenk EJ,Redekop WK,Luyendijk M, Rijnsburger AJ, Severens JL. Productivity Loss Related to Neglected Tropical Diseases Eligible for Preventive Chemotherapy: A Systematic Literature Review. PLoSNegl Trop Dis2016; 10: e000439.
  27. Congdon N, Yan X, Lansingh V, et al.Assessment of cataract surgical outcomes in settings where follow-up is poor: PRECOG, a multicentre observational study. Lancet Glob Health 2013; 1: e37–45.
  28. Ramke J, du Toit R, Palagyi A, Brian G, Naduvilath T. Correction of refractive error and presbyopia in Timor-Leste.Br J Ophthalmol2007; 91: 860–6.
  29. Fox MP, Rosen S, MacLeod WB, et al.The impact of HIV/AIDS on labour productivity in Kenya. Trop Med Int Health2004; 9: 318–324.

.

Table 1· Baseline characteristics of participants by randomization group  
 

Characteristics

 Control Group

n=375 (49·9 %)

 Intervention Group

n=376 (50·1 %)
Age, years    
 40-44, n (%) 151 (40·3%) 132 (35·1%)
 45-49, n (%) 87 (23·2%) 108 (28·7%)
 50+, n (%) 137 (36·5%) 136 (36·2%)
 Mean age (range) 47·1 (40 – 61) 47·2 (40 – 59)
Female sex, n (%) 293 (78·1%) 293 (77·9%)
*Mean height, cm (SD) 150 (7·79) 150 (8·17)
*Marital status: married, n (%) 370 (98·7%) 371 (98·7%)
Wearing glasses at baseline, n (%) 0 (0) 0 (0)
Uncorrected near vision < 6/18 in better-seeing eye, n (%) 73 (19·5%) 69 (18·4%)
Mean power of near correction in better-seeing eye, Diopters (SD) 1·62 (0·436) 1·61 (0·418)
Best corrected near vision = 6/6 (normal) in better-seeing eye n (%) 362 (96·6%) 355 (94·4%)
Mean working distance, cm (SD)
 40-59 cm, n (%) 40 (10·7%) 52 (13·8%)
 60 cm, n (%) 122 (32·5%) 122 (32·5%)
 61-70 cm, n (%) 213 (56·8%) 202 (53·7%)
*Mean time working as a tea picker, years (SD) 27·4 (8·06) 26·2 (8·39)
*Agree or strongly agree: Important to maximize income by picking as much tea as possible, n (%) 374 (99·7%) 375 (99·7%)
*Agree or strongly agree: Picking tea is main source of family income during high season, n (%) 374 (99·7%) 375 (99·7%)

SD = Standard deviation   *Two participants had missing values for each of these starred variables

 

Table 2· Effect of randomization group on change in productivity (daily weight of tea picked) from baseline
 

 

Group

  

 

N

  

Baseline mean daily productivity over 4 weeks, kg/day (SD)

 

Post-intervention mean daily productivity over 11 weeks, kg/day (SD)

  

Change in productivity, kg/day

(95% CI)

 

Difference in change in productivity, Intervention group compared with Control group, kg/day (95%CI)

 
Control 375 26·0 (3·48) 30·6 (4·77) 4·59 (4·10, 5·07) 5·25 (4·50, 5·99)

(P<0·001)

 
Intervention 376 25·0 (4·25) 34·8 (5·11) 9·84 (9·27, 10·4 )

 

SD: Standard deviation   95% CI = 95% Confidence Interval

 

 

 

Table 3: Intention to treat analysis for linear regression model of potential predictors of change in productivity

 
Characteristics Univariate analysis

(n= 751)

   Full model‡

(n= 751)
β (95%CI) P-value   β (95%CI)  P-value
Intervention group (Control group as reference) 5·25 (4·50, 5·99) <0·001 5·25 (4·60, 5·91) <0·001
Age (years) (Age effect in Control group) -0·00383 (-0·0841, 0·0764) 0·925 -0·158 (-0·242, -0·0745) <0·001
Age (years) (Age effect in Intervention group) 0·112 (0·00555, 0·219) 0·039
Group × Age Interaction 0·271 (0·144, 0·397) <0·001
Female sex 5·44 (4·80, 6·08) <0·001 4·61 (3·65, 5·57) <0·001
Height, cm -0·156 (-0·201, -0·112) <0·001 -0·0339 (-0·0840, 0·0162) 0·184
Observed compliance with glasses wear (%)* 0 ·0640 (0·0473, 0·0807) <0·001 — —
Uncorrected near vision in better-seeing eye < 6/18 1·11 (-0·0934, 2·32) 0·071 1·00 (-0·126, 2·13) 0·082
Working distance
  61-70 cm Reference Reference
  60 cm 0·582 (-0·309, 1·47) 0·200 -0·0235 (-0·741, 0·694) 0·949
  40-59 cm -1·06 (-2·53, 0·409) 0·157 -1·41 (-2·62, -0·196) 0·023
Time working as tea-picker, years 0·0543 (0·00189, 0·107) 0·042 — —
Work attendance rate during evaluation period (%) 0·106 (0·0777, 0·134) <0·001 0·0367 (0·00978, 0·0635) 0·008

 

95% CI: 95% Confidence interval

‡Including variables associated with change in productivity with significance of p<0·20 in the univariate analysis (time working as tea-picker was excluded because of its collinearity with age; compliance with glasses wear was excluded because of collinearity with group assignment). Continuous variables were centeredby subtracting the mean of the variable.

*Expressed as a proportion: Number of times glasses were worn during 7 un-announced observations carried out at work over the 11-week evaluation period.

 

Figure 1

Leave a Comment