The institutional review board at the University of Alabama at Birmingham, USA approved the study protocol. Fourteen nursing homes in the Birmingham, Alabama area, USA were sites for the study. All participating nursing homes had eye care services available to residents in that a licensed optometrist visited the facility on a routine basis to provide eye care services to residents, and when the services of an ophthalmologist were needed (eg, evaluation for possible cataract surgery), the optometrist had established referral paths with community ophthalmologists for this purpose. Written informed consent was obtained from all participants and also the resident's sponsor (a family member or state‐appointed guardian) after explaining the nature and possible consequences of the study.
Persons were eligible for enrolment if:
- they were identified by the unit charge nurse as a person who could answer simple questions about vision and daily activities;
- they were 55 years old;
- they spoke English;
- they had a Mini‐Mental State Exam (MMSE)22 score of 13, since comprehension of simple requests and questionnaire items was critical to valid outcomes assessment;
- they had a cataract in one or both eyes that caused functional problems for the resident;
- the optometrist and the ophthalmologist judged that cataract surgery was an appropriate treatment option for one or both eyes given the person's ocular and overall health status.
Residents and their sponsors were informed by the ophthalmologist that cataract surgery was an option in order to improve vision. If the resident and sponsor decided to have surgery on one or both eyes, the surgery was scheduled, and if the resident and sponsor consented to study participation, the resident was assigned to the cataract surgery group. Those who declined cataract surgery and consented to participate were assigned to the no‐surgery group. All potential enrolees had health insurance that would cover the cost of cataract surgery and a new spectacle correction following surgery, should they elect surgery. Recruitment for the study proceeded over a two‐year period and was part of a larger study on the visual status of nursing home residents.7,23
The study protocol consisted of a baseline and a follow‐up assessment, which were identical. For the surgery group, the baseline assessment was administered before cataract surgery on either eye, and the follow‐up assessment was approximately 4 months following surgery after the resident had been using his/her new spectacles for 2–3 months (new spectacles are not typically prescribed until 4–6 weeks following cataract surgery). If both eyes were undergoing surgery, the follow‐up assessment was not performed until both eyes underwent surgery. For the no‐surgery group, the baseline and follow‐up assessments occurred at times that were commensurate with the pre‐post duration for the typical surgery enrolee, except of course no cataract surgery had occurred in the intervening period.
Baseline and follow‐up assessments were as follows. A research staff member assessed distance and near visual acuity while the resident used habitual correction (or nothing if they used no correction) for each eye separately and together. Testing was carried out either in the resident's room or another private area with adequate lighting. Distance acuity was assessed with the ETDRS chart using its standard protocol and expressed as logMAR.24
Near acuity was assessed using the Lighthouse Near Visual Acuity Test (modified ETDRS) administered at 40 cm. Contrast sensitivity was evaluated using the Pelli‐Robson chart25
and its standard protocol and scored by the letter‐by‐letter method.26
The Nursing Home Vision‐Targeted Health‐Related Quality of Life Questionnaire (NHVQoL)27,28
was used to assess vision‐targeted health‐related quality of life. The NHVQoL is an instrument specifically developed for the nursing home resident population and consists of nine subscales focusing on general vision, reading, ocular symptoms, mobility, psychological stress, activities of daily living, activity/hobbies, adaptation/coping, and social interaction. Two other questionnaires were also administered: the Medical Outcomes Study Short‐Form 36 (SF‐36)29
to assess generic health‐related quality of life (both mental and physical components) and the VF‐14 to assess the visual activities of daily living.30
Scoring on all three questionnaires is from 0 (severe disability) to 100 (no disability), and all questionnaires were interviewer‐administered by trained personnel. The presence of depressive symptoms was assessed by the 15‐item Geriatric Depression Scale (GDS),31,32
a widely used screening instrument for estimating depressive symptomatology in older adults including those residing in nursing homes. The Cataract Symptom Score30
was obtained by asking residents to what extent they were bothered by five visual symptoms caused by cataract: distorted vision, glare, blurry vision, colours looking different and worsening of vision in the past month. The total score ranging from 0 (not bothered by any symptoms) to 15 (very bothered by all five symptoms) estimates the extent of visual symptomatology. The medical record was abstracted to obtain information on demographic variables (age, gender, race/ethnicity and education, which were also verified by interview), length of stay in the nursing home, current chronic medical conditions, type of cataract, and characteristics of the surgical procedure if performed.
Statistical t and Χ2
tests were used to compare demographic, medical and vision characteristics between the cataract surgery and no‐surgery groups. To compare pre‐surgery and post‐surgery vision among the eyes that underwent surgery a mixed model, which accounted for the correlation of eyes within a single patient, was used. Paired t tests were used to compare baseline and follow‐up vision, health‐related quality of life, depression and cataract symptoms within the surgery and no‐surgery groups, whereas analysis of covariance (ANCOVA) was used to compare changes in these characteristics between groups. The dependent variable for these ANCOVA analyses was the follow‐up measurement and the primary dependent variable was surgery versus no surgery; the associated baseline measurement for the dependent variable was also included as an independent variable. p Values
0.05 (two‐sided) were considered statistically significant.