1.1 Purpose of Research
The study we present is a new analysis of transition data with factors by using Markov chain. Select the factors to build a statistical model and will pay more attention. In the cohort study we will follow the same participants in each time echo however the
information they provide will dependent. Today we add a Markov property assumption to solve difficulty in the analysis of dependent data. Conditional on covariates the process of each person will be a Markov chain to solve difficulty in the analysis of dependent data.
The new strait in the analysis is that we need an “enough” sample size. In this example with many factors then the size of subgroup is small not mention to the continuous factors.
Our method is to break the continuous data to categorical data simplified the level of factor and use homogeneity and stationarity tests to pick up the influential factors.
1.2 Background
Vision composition
The visional system consists of retina of eyes which connected optic nerve to the visional center of brain. The maculopathy on the retina is against to the pupil with 5.5 mm larger than the pupil. If we look straightly then the maculopathy can controls 20 degree of viewpoint. Highly sensitive visional cells are located at the maculopathy although its area occupied the all retina is just 2%. The visional center uses more then one half of cells to analyze the information by it received.
The vision consists of retina and the maculopathy is not only the geographic center but also the center of visional center. Once the maculopathy has pathologies then the vision is
effected by it even loosed one's sight.
1.3 Literature review
The previous researches in the Beaver Dam Eye Study [7-17] provide some statistical method but most of them are focus on the single events rate such prevalence rate or
procession rate et al. Our method in this paper not only solve the dependent data but also provide a global view to understand how a factor/factors act on the ARM.
These risk factors were chosen because of a strong relation with age-related
maculopathy in previous studies. In the Beaver Dam Eye Study, smoking was related to the prevalence of age-related maculopathy [12], heavy drinking and hypertension were
associated with exudative macular degeneration, a lesion that defined late age-related maculopathy [14, 16], and serum cholesterol was inversely associated with age-related maculopathy [10]. Vitamin use was found to be associated with the incidence of
age-related maculopathy in a clinical trial [1]. Definitions of these confounding variables have been described in detail elsewhere [2, 12, 14 and 19]. In brief, a subject was classified as a current smoker if he/she had smoked more than 100 cigarettes in his/her lifetime and had not stopped smoking; as a former smoker if he/she had smoked more than this number but had not smoked within the last year prior to the examination; and as a nonsmoker if he/she had smoked fewer than 100 cigarettes in his/her lifetime. A current heavy drinker was defined as a person consuming four or more servings of alcoholic beverages daily, a former heavy drinker had consumed four or more servings daily in the past but not within the last year, and a non-heavy drinker had never consumed four or more servings daily on a regular basis. A person was classified as a current vitamin user if he/she had taken at least one vitamin per week in the month prior to the examination; as a past vitamin user if he/she had ever regularly taken vitamins at least once a week but not within the last month; and as never using vitamins if she/she never took vitamins regularly. Hypertension was defined as
a systolic blood pressure of 160 mmHg and/or a diastolic blood pressure of 95 mmHg and/or a history of hypertension using antihypertensive medication at the time of the examination. We adjusted for these potential confounding variables in each model.
Measurements of risk factors were taken at each examination; however, multivitamin use and cholesterol level were not available at the 10-year follow-up. In the following analysis, we use the 5-year multivitamin use and cholesterol level as the 10-year measurements.
The possible reasons for nonparticipation include death, moving out of the area, and refusal [9, 11 and 13] .Comparisons between participants and non-participants at all three examinations have been presented elsewhere [9, 11 and 13].
1.4 Procedures
Procedures for obtaining and evaluating photographs of participants' eyes have been described elsewhere [9, 17]. At each examination, 30° color stereoscopic fundus
photographs were taken of both of each participant's eyes. Preliminary and detailed grading was then carried out on the fundus photographs to determine the presence and severity of specific lesions associated with age-related maculopathy, including largest drusen size, most severe drusen type (in order of increasing severity: hard distinct drusen, soft distinct drusen, soft indistinct drusen, and reticular drusen), increased retinal pigment, retinal pigment epithelial depigmentation, exudative macular degeneration (retinal pigment
epithelial detachment, subretinal hemorrhage, subretinal fibrosis), and geographic atrophy.
In this reporter, we adopt 6-level scale. Experienced graders used the photographs to evaluate the severity of lesions of ARM, which were graded on a 6-level scale, such as 10 20 … 60 [14].
The scale will be re-classified to three levels (the detail is the following definition), in order to increase severity: level 0 = disease free, level 1 = early ARM and
level 2 = late ARM. Results presented here use each individual's ARM level in the worse eye. [6]
Define 1.1 the states of ARM
Level 0: disease free if 6-level=10 Level 1: early ARM if 6-level=20/30/40 Level 2: late ARM if 6-level=50/60