DELIVERY OF CARE TO CANCER PATIENT

Abstract

In today’s world, the military physicians continually deploy, leaving their patients’ without a primary provider and making it difficult for them to practice continuity of care. This is especially true in the case of cancer treatment.

As the seriousness of the disease grows, the importance of continuity also increases. Unfortunately, the (2003) review, found cancer care in the United States fragmented and without coordination. The results of this study have shown a moderate level of continuity for Irwin Army Community Hospital’s (IACH) cancer population, specifically for the breast cancer patients. While the hospital should be pleased with this finding, it still needs to improve its level of continuity. One approach to emphasizing continuity of care is to evaluate it periodically. If continuity is important, then how it is measured is also significant.

The study begins by attempting to compare, analyze, and recommend the most stable model to measure continuity. The objectives for conducting this study are as follows: first, to determine if the delivery of care of Irwin Army Community Hospital (IACH) provides optimal continuity of care; second, to determine which variables influence continuity of care; and third, to determine which variables isolate those patients in the greatest need of continuity.

The findings of the study expect to demonstrate that the current process for continuity might be fragmented and inefficient because the hospital’s main priority is supporting the active duty population. At times, this, and the fact that the hospital is short of physicians, influences the staff’s ability to see the other beneficiaries.

TABLE OF CONTENTS

INTRODUCTION…………………………………………………………………..……3

Conditions which prompted the study………………………………..………13

Statement of the management question.……………………………………14

Literature Review………………………………………………………………15

Purpose…………………………………………………………………………27

Hypothesis………………………………………………………………………27

METHODS AND PROCEDURES…………………………..……………………………28

Subjects and Events……..……………………………………………………28

Study Design…..……………………………………………………………….30

Data Analysis……..……………………………………………………………31

Results…….……………………………………………………………………33

Limitations………………………………………………………………………37

Ethical Considerations…………………………………………………………37

DISCUSSION AND CONCLUSIONS…………..……………………………………39

LIST OF REFERENCES………………………………………………………………43

APPENDICES….………………………………………………………………………49

Appendix A……………………………………………………………………..49

Appendix B………………………………………………….………………….50

Appendix C……………………………………………………………………..51

Appendix D……………………………………………………………………..53

Table 1………………………………………………………………………….54

Table 2………………………………………………………………………….55

Table 3………………………………………………………………………….56

Table 4………………………………………………………………………….57

Table 5………………………………………………………………………….58

Table 6………………………………………………………………………….59

Introduction

For the year 2000, those who died of cancer represented 23% of all deaths in the United States ( 2001). The top ten cancer types vary for men, women, and children, but the chances for developing cancer are high in all groups ( 2003). In addition, as the population of older Americans increases, the number of people with a cancer diagnosis will double, increasing the demand for resources. As this disease’s population growth continues, the strain on consumption (treatments for cancer, control medications) will continue to rise (, 2003).

(2001) reported several barriers to quality care for cancer patients. These barriers include adequate cancer training for providers, poor management of cancer-related symptoms and lack of timely referrals. Additionally, untold numbers of patients did not receive appropriate medical care because their geographical location was not conducive to receiving treatment. (, 2001).

These conditions resulted in the loss of public confidence in the system. The National Cancer Policy Board (NCPB) responded by doing a review of the effectiveness of cancer services. They found an “ad hoc and fragmented cancer care system that did not ensure access to care, lacks coordination, and is inefficient in its use of resources” (, 2003). This prompted Irwin Army Community Hospital (IACH) to investigate continuity of care for their cancer patients as a way of improving patient treatment.

Continuity of care, specifically for cancer patients, in military health care has been included in many efforts of improving treatment. It is one of the benefits reaped from the establishment of an Internet Tumor Board, a telemedicine system that is web-based to allow medical specialists to coordinate the care of cancer patients in the Asia-Pacific area. The group that established this system had trouble with their patients being sent from one location to another, referring patients to the center facility for major treatments then handing them back to the local hospitals for continued treatment. Using this, they were able to eliminate potential for discontinuity that came from the transfers, and even from the frequent transfers of the military doctors themselves. (, 2000) This, and other researches and projects brought through agencies such as the U.S. Army Breast Cancer Research Program have sought ways in which treatment and care can be improved.

This then is an important component in providing quality patient care. Research in continuity of care suggests a strong correlation between higher continuity and improved patient satisfaction. Better continuity is associated with fewer hospitalizations, improved compliance, and lower emergency room use (, 2001). Unfortunately, recent statistics reveal cancer patients are not receiving the best care due to inconsistent continuity (i.e. receiving care from multiple physicians).

The first step in this task began with settling on a definition of continuity. (2001), article, aptly defined continuity in the context of quality health care as “good to the extent it increases likelihood of desired health outcomes and consistent with current professional knowledge while also providing patients with appropriate services in a technically competent manner, good communication, shared decision-making and cultural sensitivity” (p. 6).

(1980) defined continuity as: An uninterrupted succession of events or the existence of a mechanism to bridge parts to an event. In medical terms, the successions of events are episodes of an illness, and continuity is the means to bring the episodes together to evaluate the illness. (p. 123)

With these definitions, we see that the primary intent of continuity is to improve follow–up of patient’s problems and facilitate efficiency in diagnostic workup and management. The information about the plan of care passes from one visit to the next (, 1980). Continuity of care is dependent on provider consistency, or involvement with a limited number of consistently available providers aware of the patient’s medical history. Thus the components of continuity include:

· Medical care consistent with current knowledge

· Competency

· Appropriate services

· Mechanism that brings multiple episodes of treatment together to evaluate the illness

· Good communication

· Shared decision making

· Cultural sensitivity

Using these components of continuity of care, the researcher will explore the scope of cancer services provided by IACH providers. This will include a review of the standards of practice accreditations that ensure quality cancer care, such as putting emphasis on the reliability of test data results which gives research its merit.

Currently though, IACH has earned the American College of Surgeons (ACoS) Commission on Cancer (CoC) accreditation in the Hospital Associate Cancer Program category. This means on site services are limited but are available by referral (See Appendix A). According to (2003) this also means that IACH’s cancer patients will receive:

· Quality of care close to home.

· Comprehensive care offering a range of state of the art services and equipment.

· Multi-specialty, team approach to coordinate the best cancer treatment options available.

· Access to cancer related information, education, and support.

· A cancer registry that collects data on cancer type, stage, and treatment results and offers lifelong patient follow-up.

· Information about clinical trials and new treatment options. (p. 2)

In order to maintain accreditation, IACH must undergo an on-site review every three years, where continuity can be measured alongside the listed factors. To examine continuity of care in a given patient population, tools such as the Modified Continuity Index (MMCI) and others can be used.

Continuity of care is a necessary part of the health care process; it leaves a positive impression on patients, and increases satisfaction. With increased satisfaction, beneficiaries will continue to use the facility for their entire health Care needs. But as military health care moves to the Next Generation of TRICARE contract, a military facility unable to provide a modern standard of care or deliver a perceived quality of care, equal to that provided in the civilian community may experience a loss of enrolled members (, 2000).

Conditions Which Prompted the Study

IACH underwent a re-organization to the military’s health plan TRICARE in 1988. This resulted in the closure of the Department of Medicine. IACH no longer had the capability or the staff to see other than active duty soldiers and dependents within their catchment area (See Appendix B). All retirees and their families felt the impact of these decisions in 1998. By the end of 2000, this affected the cancer patients, forcing many to find a physician in the private sector. Unfortunately, some of these patients did not find another provider. According to , (March 15, 2004), the Tumor Registrar at IACH, “It is this group and even some of those with outside providers that show up multiple times in IACH’s emergency room (ER) when they are very sick.” This leaves a group of beneficiaries without a central source of care.

In October 2004, TRICARE, the military health plan, will transition to the Next Generation of TRICARE contract although the basic benefit structure (TRICARE Prime, Extra, Standard, and Plus) will remain unchanged. Some of the benefit plan responsibilities will transfer from the TRICARE contractor to the military treatment facility (MTF) Commander. This change provides incentives for the MTF to adopt best practices, provide measurable performance in customer service, quality of care, and access to care (, 2003).

The MTF, funded on historical data, will also receive the revenue paid in the past to the contractor. The MTF will then pay the contractor for the services Prime enrolled beneficiaries receive in the network. This provides the MTF with the flexibility to determine whether to provide in-house or network services based on cost effectiveness and the need of beneficiaries in the area.

Currently, IACH provides health Care services to a population of over 32,968 with a budget of .6 million for the 2002 fiscal year. IACH authorized staff comprises 266 military, 360 civilians, and multiple contract personnel. The shortage of providers described here is incremented by IACH’s support on the war on terrorism by the deployment of their medical staff.

IACH is consciously aware that the shortage of physicians, inability to see all enrolled beneficiaries, and the changes in the TRICARE contract affects continuity of care. A review of the current researches conducted in this field has shown that there are few who focus on the topic of continuity in this context. Rather, like those that are included under the Department of Defense’s Congressionally Directed Medical Research Programs, most of them only list continuity under the benefit of primary care improvement, or a requirement of it.

This study will help IACH measure the continuity of care for their breast cancer patients under these considerations. Additionally, it should help determine the details of the factors that impact continuity of medical care.

Statement of the Management Question

This study will attempt to answer the questions identified from the literature associated with continuity of care and then apply them to the cancer patients at IACH. In particular, the researcher seeks to reply to the following questions:

Research Question 1: Do IACH providers offer continuity of care to their breast cancer patients?

Research Question 2: To what extent do outpatient visits, readmissions, and ER visits predict MMCI scores?

Research Question 3: Are there any relationships between MMCI scores with age, gender, race, beneficiary status and length of disease?

Research Question 4: Do the participants’ MMCI scores differ by stage and site of cancer?

Literature Review

To evaluate the study methods and measurement techniques which have been employed in the study of continuity of care, ’s includes those that used age at diagnosis, gender, and length of the relationship and stage of cancer as variables that might influence continuity of care. These were used to answer the question of the relationship between continuity and emergency room (ER) visits and those between cancer type/stage and frequency of visits. But all of these studies do not agree on the definition and role of continuity. (pp. 1-19) So from this we see that the study of continuity involves visiting patterns and concentration of providers

On the other hand, (1994) felt that “continuity of care is established when a patient’s physician is able to understand interrelationships” (p. 14). In other words, the patient’s feelings, past medical history, signs, symptoms, treatments, and responses to the treatments all need to be included to define or understand continuity of care.

Unfortunately, delivery of care to cancer patients is usually complex and uncoordinated, resulting in patient handoffs that delay care and decrease the quality of care over the continuum. In turn, these developments are associated with bad health outcomes or even death (, 1999).

Regarding this, (1975) proposed that “there are certain medical conditions that need a higher degree of continuity. For example, a patient will require additional follow-up appointments when there is a chronic problem than an injury requiring sutures. The age of the population may also be important, as older age groups tend to have more severe long-term health problems. The high mobility of the American population may render the goal of continuity unattainable.” (p. 141)

The importance of continuity of care increases with the seriousness of the condition. When a lack of continuity occurs, it is disruptive and stressful for the patient with chronic conditions. Patients perceive they receive better care if they have continuity, creating a higher level of trust (, 2002). When seen by their regular physician, patient satisfaction increases with physician performance ( 1996). Positive patient-provider relationships have shown to affect certain health behaviors such as lowering emergency room visits and even hospitalizations ( 2001). Long-standing physician-patient ties foster less expensive and less intensive medical care (, 1996).

Efforts to improve continuity may improve the quality and outcomes of care ( 2001). (2003) found in their literature review that “greater continuity is associated with reductions in hospitalizations, improvement in follow-up, increase in patient satisfaction, promotion of compliance with recommended care, and a reduction in duplication of tests” (p.992).

(2004), in their literature search, also found evidence that “continuity of care is associated with a decreased likelihood of future hospitalizations and less ER use” (p. 36).

(1975) proposed that continuity is a must for patients with chronic problems. (2002) suggested the importance of continuity increases with the seriousness of the condition. Other studies demonstrate that continuity decreases duplication of tests, ER visits, hospitalizations, which decrease the cost of care. This information substantiated the value to further investigate continuity of care. It thus became important to discover how other continuity studies’ formulated their research design. Unfortunately, there is not any continuity of care research from military facilities to use as a model.

The following three studies either helps define variables, narrow definitions; show categorizing within a variable or present design possibility for this research.

First, (1987) started by defining measures of continuity as: the relationship between a patient and his or her primary provider and the generation of a meaningful numerical value that is easily understood. They collected data for 201 patients. The patients included made 1,154 visits to physicians over a two-year period. Fifty-nine percent were female; the average age was 28 with a range 1 year to 80, sixty-five percent had an assigned provider and twenty-six percent of the patients had all encounters with the same physician. The result section included a COC, MCI, and MMCI scores for each patient (Appendix C). The findings substantiate that the use of UPC calculated scores of more than twice those calculated for COC. The COC is sensitive to number of physicians seen and the MCI is results that may mislead users ( 1987).

Second, a study by (2003) collected data on deceased cancer patients (8,702) with a confirmed pathological cancer report. They limited their research to office visits for the last 180 days of the subjects’ life. The MMCI measured continuity of care as an independent variable and ER visits (total count of ER visits during each patient’s survival time) as the dependent variable.

They used descriptive statistics, followed by negative binomial regression analysis with adjustments for varying survival times to assess the association between continuity and the total number of ER visits. Then variables associated with ER use in the bivariate analysis at a 0.1 level of significance and associated with continuity were included in the multivariate regression, along with the demographic variables (e.g., sex by age, sex by cancer death).

Finally, their subsequent modeling employed manual backward elimination to develop the final model of total emergency room use and association of provider continuity of care. The skewed distribution of their continuity scores forced them to categorize continuity as being low (scores < 0.5), medium (scores 0.5 to < 0.8), and high (scores 0.8 and greater).

The majority of their patients were 65 years of age or older (72%), female (45.1%), male (54.9%), the most common cancer was lung (26%) followed by breast (10.2%), prostrate (9.2 %), colorectal (8.5%) and all others (53.1%). The continuity scores ranges included low (0.02-0.47, (8.1%)), moderate (0.50-0.79, (35.6%)) and high (0.80-1.00, (56.4 %)), ( 2003).

And the third under the group of studies we’ve been following is by (2000), and was concerned with the effect of continuity of care on ER use. It is of importance because it compares UPC and MMCI scores but also provides examples of definitions for variables, narrows the definition of visits, categorizing within a variable when needed and uses ER visits as a dependent variable.

Their analysis was performed by using the MMCI index as their main continuity measure. However, since the UPC index is commonly used, they ran an additional analysis in place of the MMCI. They defined visits as office or clinic visits to a physician, nurse practitioner, or a physician assistant. They excluded ER visits because it was their dependent variable. Additionally, they excluded special procedures and visits to other providers such as chiropractors, optometrists, and psychologists. They categorized each patient’s ER visits into three levels (no ED visit, 1 ED visit or multiple visits). This categorization allowed them to measure the effect of a single ER visit to multiple visits in the same model. They did not examine ER visits as a continuous variable because of the extreme bias and kurtosis of the data violated the assumptions for linear regression.

In their main analysis, they measured the association between MMCI and ED visits by using a polychromatous logistic regression, with a 3-level categorization (no ER visit, 1 ER visit or multiple visits) of ED visits as the dependent variable. They used a categorical data model because the different levels of the predictors did not meet the assumption of proportionality of odds. They first ran the model with only continuity as the predictor variable. They then ran the model after controlling for the number of office visits during the study.

Their previous analysis showed the number of office visits to be a strong negative confounder of the association between continuity and hospital use (i.e., office visits are positively associated with the dependent variable but negatively associated to the independent variable, so this would result in an apparent negative association between the independent and dependent variables. They then ran the model with continuity and all control variables forced into the model to determine the effect of continuity.

Finally, they ran the model using a stepwise technique, to determine the best predictor model. For the stepwise technique, they included predictors associated with ED visits from the univariate analysis with a P < .25 or the variables considered clinically relevant. Stepwise selection showed P < .05. They then added variables back to the model if they changed the co-efficient of the other variables by more than 15%. The final model showed good predictor accuracy (c=0.81). (pp. 334-335)

The preceding sections of this literature review establishes that continuity of care results in better health outcomes, helps coordinate care, increases trust, and reduces hospitalizations. Provider and emergency room visits also decrease which in turn lowers health care costs. This prompted the researcher of this study to evaluate the best measurement for continuity of care.

The (1987), “study considered three models used in the past to measure continuity of care. They found each of these continuity indexes did not truly represent the relationship of total number of visits to the total number of providers” (p.167). Building on strengths of each measurement, the authors developed a new model that truly represents the total number of patient visits to the total number of providers.

The author explored measure to determine which is best suited to use in this research. The first measure, usual provider of care (UPC), oversimplifies the relationship between patient and provider. UPC is simply a ratio, determined by limiting the visits to highest number of visits to one provider divided by the total number of providers. The UPC produces a measure that ranges from zero to one. The zero represents all visits to different providers and one represents all visits to the one provider.

UPC = V/Pr (1)

V = Total number of Visits

Pr = Total number of Providers

For example, if a patient has eight visits, (four visits to one physician and four visits to four other providers), the calculated UPC is 0.8 (4/5). The problem with the UPC concept is that it does not accurately account for the total number of visits. (1980) and (2003) have substantiated in their research that the UPC does not emphasize the number of different doctors seen.

The second measure they examined was continuity of care (COC).

[(Pr)2 + (Pr)2 + (Pr)2 + (Pr)2 + (Pr)2)-V] (2)

[V (V-1)]

V = Total number of visits

Pr = Numbers of visits to provider

This formula attempts to account for visits to individual physicians. For example, if a patient has a total of eight visits, four visits to a specific physician and the remaining four, each to different providers, COC would be 0.214 while the UPC would be 0.8. The COC also produces a measure that ranges from zero to one. The zero represents all visits to different providers and one represents all visits to the one provider.

The UPC does not accurately account for total number of visits. Additionally, it falls rapidly with a large number of providers.

The third measure is the modified continuity index (MCI), which tries to provide sensitivity to the total number of visits and providers (1984). The formula for MCI is as follows:

MCI=1-(Pr of providers / [V of visits + 0.1]) (3)

V = Total number of visits

Pr = Numbers of visits to provider

This measure also generates a continuity score of zero to one. The zero represents all visits to different providers and one represents all visits to the one provider. Again, using a patient with eight visits, four visits to the same provider and the remaining four to four different providers; UPC=0.8, COC=0.214 and the MCI=0.38. While the MCI is less sensitive and less misleading, it still does not truly reflect the relationship to the total number of providers to the total number of visits.

The shortcomings of previous measures for continuity prompted (1987) to develop a new more meaningful measure, the MMCI. This measurement tool is derived from the modified continuity index. The MMCI produces a measure that will also range from zero to one. The zero represents all visits to different providers and one represents all visits to the one provider.

The formula is as follows:

MMCI = 1-(Pr / [V  0.1]) (4)

1-(1 /[V + 0.1])

V = Total number of visits

Pr = Numbers of visits to providers

Again, using a patient with eight visits, four visits to the same provider and the remaining four to four different providers; the UPC = 0.8, COC=0.214, MCI=0.38 and the MMCI of 0.44. The MMCI measure is not overly sensitive to the large number of providers and allows intuitive interpretation of continuity regardless of number of providers.

Before deciding to use a measurement, it is important to check the instrument’s validity or the extent to which a test measures what it purports to measure. If a patient has two visits to two different providers (See Appendix C, row 2), the resulting MMCI is 0.10. If you compare the results of the other measures UPC=0.50, COC=0.00, MCI=0.05 to the MMCI, it becomes clear that the MMCI is less misleading. It also accurately represents a small number of visits. This also establishes the preciseness of the index, which is one of the components of reliability.

Appendix C (Row 4) also demonstrates the value and validity of the MMCI. If a patient has a total of three visits, two visits to one provider and one to a different provider, the score of the MMCI=0.51, UPC=0.67, COC 0.33, and the MCI=0.35. The MMCI comes the closest to truly representing the total number of visits to total number of providers. The COC and MCI scores give the impression of very low continuity and the UPC is misleading.

Reviewing the previous sections, this study then used the MMCI.

On that note, this study will employ the Modified Continuity Index (MMCI) to examine continuity of care in this patient population. The researcher feels that previous measurement tools did not accurately represent the connection between the number of providers seen and total number of visits. The MMCI is the most consistent true measure and the most predominant model currently found in the literature. It accounts for the degrees of dispersion among different providers. Continuity for this study is a MMCI .51 or greater.

Purpose

This study was conducted with the following three objectives in mind. First, it will determine if the delivery of care for cancer patients at IACH provides optimal continuity of care.

Second, to determine which variables influence the continuity of care. And third, to determine which variables isolate those patients in the greatest need of continuity of care.

The findings will demonstrate that the current process for continuity of care may be fragmented and inefficient. This is because IACH’s main priority is supporting the active duty population. At times, this influences the staff’s ability to see other patients. Also the hospital and the military are below strength in number of physicians. Lastly, patients often opt to see the first available provider instead of waiting for their assigned physician.

Hypothesis

H0: The current process for care of breast cancer patients at IACH provides continuity of care.

H1: The current process of care for breast cancer patients at IACH does not provide continuity care.

Methods and Procedures

The purpose of this section is to review the description of the participants in this archival study, present the measures and procedures used to collect the data, and discuss the proposed treatment of the data.

Subjects and Events

The majority of IACH’s beneficiary population resides within the Fort Riley, Manhattan and Junction City zip codes. The use of the ACTUR database limited the participants to this area. For inclusion in the research, each subject was an IACH beneficiary at time of the initial cancer diagnosis. Additionally, patients must have had at least four outpatient visits. The consensus of research related to continuity of care, supports the inclusion of patients with four or more visits.

IACH cancer patients’ data came from ACTUR database, the Composite Health Care System (CHCS), (for services within IACH) and the Military Health Services Management and Analysis Reporting Tool (M2), (for services outside IACH). This study selected only specific data related to patients with the diagnosis of cancer. Appointments for tests (X-ray, MRI, CT, labs) or treatments (chemotherapy) were not included.

The data collection included: diagnosis, date of diagnosis, age at diagnosis, stage of disease at diagnosis, and current documented “evidence” or “no evidence” of disease. The demographic information included: beneficiary status (CHCS), age at diagnosis (ACTUR), gender (ACTUR), and race (ACTUR). The Composite Health Care System and M2 provided hospitalizations (number of separate admissions), number of providers, visits to a physician and to the ER. Each subject had a MMCI Index calculated. The continuity measure indicates the degree of continuity from January 1995-February 2004. The MMCI value is a measure ranging from zero to one.

For the purposes of this study, MMCI scores .51 and greater represented continuity. Visits are defined as outpatient visits to a provider but do not include visits for testing (labs, diagnostics (x-ray, CT, MRI) or chemotherapy) nor does it include emergency room visits in the MMCI calculations. A provider is defined as licensed medical doctor, nurse practitioner or a physician assistant.

The accuracy of the data collected is an important aspect of any study. To ensure the data is consistent, accurate, and precise (reliability), the researcher and the tumor registrar, reviewed it for accuracy.

Study Design

The review of past studies helped formulate the design for this research. This is a retrospective study. Irwin Army Community Hospital’s Tumor Registrar granted permission to use the ACTUR database to glean information for this analysis.

The ACTUR database provided cancer patients’ length of years with the diagnosis (continuous), staging (See Appendix D for coding), age at diagnosis and follow-up testing, and “evidence” (recorded as 1= no evidence of cancer and 2 = evidence of cancer). The CHCS database provided beneficiary status (20=active duty/retirees, 30=all dependents), outpatient and ER visits, total number of providers, and IACH hospitalizations for each subject. The Military Health Services Management and Analysis Reporting Tool (M2) database provided any appointments to outside providers and hospitalizations to other facilities. The data went into one file. With this merger, the influences on continuity emerged.

Data Analysis

SPSS 11.5 analyzed the data. The use of descriptive analyses (i.e., summary statistics) of the variables helped determine their characteristics, estimate the proportion of participants that have those characteristics, discovered the associations among the different variables and measured cause and effect among the variables ( 2001). Cause is defined as the variables that cause effects or influence continuity. Descriptive statistics will analyzeed all subjects while regression was limited to the breast cancer types group with 4415 or more subjects. This was the only group with 15 or more subjects. (1996) found, for a reliable regression equation, 15 participants per predictor are needed” (p. 72).

Research question 1: Does IACH providers offer continuity of care to their cancer patients?

Each subject had a MMCI score calculated. The definition for continuity of care is a MMCI score of .51 or greater. The visits used to calculate the MMCI included visits both within IACH and outside the hospital from January 1995 thru February 2004. The visits were limited to a licensed medical doctor, nurse practitioner or physician assistant. The visits did not include treatments, diagnostic testing or therapy. Both provider and visit data were checked and rechecked for accuracy by the researcher and the tumor registrar. To answer this research question, the frequencies of the MMCI scores .51 or greater determined continuity of care.

Research question 2: To what extent do outpatient visits, readmissions, and ER visits predict MMCI scores?

A multiple regression examined the extent to which the independent variables of outpatient visits, readmissions, and ER visits predicted the dependent MMCI scores. The assumptions of regression included (multicollinearity (Some or all variables are highly correlated), linearity (the collection of data can be described as a straight line), and homoscedasticity (data evenly dispersed both above and below the regression line), (2001). (1966) explain, “it is common in a multiple regression analysis to have a high degree of correlation between two or more explanatory variables (independent) that it is impossible to measure accurately their individual effects on the explained (dependent) variable” (p. 66).

Research question 3: Are there relationships between MMCI scores with age (continuous), race (1=white, 2=other), gender, beneficiary status (20=active duty or retired, 30=dependents) and length of disease (continuous)?

“Evidence” and “no evidence” (determined by yearly follow-up testing) of cancer established any changes in severity of the diagnosis. Specifically, the use of Pearson correlation coefficient (determines relationships between variables) and point-biserial correlations (dichotomous variables that represent a natural dichotomy such as evidence and no evidence) determined correlations for continuous and dichotomous variables, respectively.

Research question 4: Do the participants’ MMCI scores differ by stage (See Appendix D: codes) of cancer?

The assumptions of ANOVA, normality (the normal distribution of the mean and the standard deviations) and equality of variance (measures the dispersion of the values determined by taking the square root of the variance) assessed the difference on MMCI scores by stage (See Appendix D). Stage seven (distant metastases) was treated the same as other categories of stages. The total number of outpatient visits, used in calculating MMCI, helps control for disease severity. The outpatient visits help interpret the statistical relationship between the MMCI score and stage. Results from other studies, show the total number of outpatient visits, is associated with severity of disease, increase in ER use, and hospitalizations. Groups that exceed 5% (6) of the population of this study (118) will be used in the calculations (, 2001).

Results

One hundred and eighteen patients, 41 (34.7%) males and 77 (65.3%) females, were initially included in this study. Eighty-eight (74.6%) were identified as Caucasian and 30 (25.4%) as “other.” Of these patients, 42 (35.6%) had active duty or retired status, while 76 (64.4%) had dependents to include children. The mean MMCI score for this group .59, which by definition signifies some continuity (continuity is >.51). Table 1 presents the minimum, maximum, mean and standard deviations by age, years of cancer, number of visits, number of providers, ER visits, hospital days (HD), Total Providers (TP), Total visits (TV), MMCI and evidence of cancer. Table 2 presents the frequency and percent of staging.

Of the initial sample, 44 women with breast cancer were examined to observe relationships among variables. Thirty-four (77.3%) were identified as white and 10 (22.7%) as “other.” Table 3 presents the minimum, maximum, mean and standard deviations by age, years of cancer, number of visits, number of providers, ER visits, HD, TP, TV, MMCI and evidence of cancer. Table 4 presents the frequency and percent of staging. In terms of evidence of cancer, 2 (4.5%) did have evidence while 42 (95.5%) were still cancer free. The mean MMCI score for the breast patients was.58.

Research questions

The first research question examined if IACH provided continuity of care to their breast cancer patients. Continuity of care was defined as an MMCI score at .51 or above; Table 3 shows that the mean MMCI score for breast cancer patients was .58 (SD=.22). Therefore, continuity of care was given to these patients.

The second research question examined to what extent outpatient visits (OPV), readmissions (RA), and ER visits (ERV) predicted MMCI scores in breast cancer patients. In terms of assumptions, standard scores were calculated for MMCI scores, and assessed for outliers, where there were none. Multicollinearity was assessed with the variance inflation factor statistic (VIF). VIF scores above 10.0 are problematic, and the VIF’s for the predictors were all below 2.0. Linearity and constant variance were assessed by plotting the residuals verses the predicted values, where these assumptions were met.

The three predictors (OPV, RA, ERV), accounted for 17.1% of the variance in MMCI scores, and the model was marginally significant, F (3, 40) = 2.75, p = .055. Table 5 shows that visits were a significant predictor of MMCI scores.

The third research question examined the relationships between MMCI score with age, race (Caucasian =1, other=2), years with cancer, and years seen. Table 6 shows the correlation coefficients, where race and years with cancer were statistically significant with negative correlations. The negative relationship with race indicates that “other” races tended to have higher MMCI scores than Caucasians. The negative relationship of MMCI and years with cancer indicates that as years of cancer increases, MMCI scores decrease.

The fourth research question examined if differences existed in MMCI scores by stage (0 vs. 1 vs. 2/3). Stages 2 and 3 were collapsed because there were too few (only 2) stage 2 patients. The assumptions of normality were met, and Levene’s homogeneity of variance was non-significant. The ANOVA was not statistically significant, F (2, 41) = .01, ns. All three groups had similar MMCI scores: stage 0 (M=.57, SD=.15), stage 1 (M=.58, SD=.23), stage 2/3 (M=.59, SD=.23).

Limitations

First, this study is limited to cancer patients treated at IACH. The results are only applicable to the cancer patients in this one small military facility. The analysis only included those patients with four of more visits. All persons with one visit would be classified as having a perfect MMCI score, which is not an accurate representation.

The small number of subjects in this study may not show associations that might be found in a larger study. Additionally in some cases, the use of the categorical data optimized manageability but also sacrificed the details of that data. On the other hand, this type of data forms solid scientific evidence.

After establishing facts with the categorical data with this study, subsequent studies, such as qualitative researches, can be done to describe and support the details of its findings.

Ethical Considerations

The ethical conduct of this study was extremely important to ensure that the benefits of conducting this research was not overshadowed inappropriate data collection or disclosures. The identities of the cancer patients were kept confidential as in keeping with the guidelines stated within the Report which follows the principles of (1) the respect for persons, (2) beneficence, and (3) justice. (, 2000)

According to the report, participants in a research must be treated as autonomous agents with the capability of deciding and making choices for themselves. The researcher is then accountable for participants with diminished autonomy, such as those with physical or mental incapacities. The participants should not come to any form of harm under the study, and the researcher should extend considerations for good outcome from the study to include the benefits to the participants themselves, science, and humanity.

Any participant should have voluntary informed consent to be in the study, and should have been informed of all procedures, purposes, and risks inherent therein, as well as being able of legal capacity to understand the information entailed by the research.

Discussion

As previously stated, the military health care setting is one of the more difficult settings in which to practice continuity. In today’s world, the military physician faces being deployed, leaving their patients’ without a primary provider. One approach to emphasizing continuity of care is to periodically evaluate it. If continuity is important, then how it is measured is also significant. The literature review substantiated the strengths of the MMCI. But this study additionally demonstrated the strength of the MMCI to the other measurements in use.

The mean MMCI score for this study average was .59 and .58 for the breast patients. For comparison, the study with terminally ill cancer patients found a mean MMCI score of 0.78. This indicates that while this group has a moderate measure of continuity, there is room for improvement.

Not only is it important to measure continuity but to also discover the factors that impact it. Once these are isolated then a plan can be in place to increase continuity. The outpatient visits, readmissions and ER visits model for the breast cancer patients was only moderately significant. But, the number of outpatient visits proved to be a significant predictor of the MMCI scores. This is supported by the work of (YR) and (YR).

While the ER visits and readmissions were not significantly predictive of MMCI in this research, other studies by found ER visits are a predictor of MMCI scores. Studies by studies found the number of readmissions are also a predictor of MMCI scores.

Interestingly this study did find that races other than Caucasians had higher MMCI scores. This result has not been reported in any of the other studies.

Research by has found as the years of cancer increase, the MMCI scores decrease. The findings of this study further support this literature.

Studies by suggested that patients with chronic diseases require a higher degree of continuity. In this research, MMCI scores did not show a difference by stage. This is similar to the study where there was no significant association between continuity of care and stage of cancer at diagnosis.

The results of this study have shown a moderate level of continuity for IACH’s cancer population and specifically the breast cancer patients. The literature review established the importance of continuity increases with the seriousness of the disease. Unfortunately, 2003 review, found cancer care fragmented and without coordination. Irwin Army Community Hospital might be able to increase the level of continuity with additional health care providers. However, this option is not likely to occur. Priorities demand providers fulfill their main mission, which is to deploy with the soldiers, ensure the medical fitness of troops that mobilize and demobilize, and care for the injured soldiers returning from the war in Iraq. An alternative is substituting case managers to follow the cancer patient population. They coordinate care and follow up with patients. (1998) wrote “case management is a primary means to provide continuity of care across multiple institutions and practitioners” (p. 151). While not the best answer, it might maximize continuity. Another option is to refer patients to the local referral facilities. Unfortunately, they also have capacity limits. According to (2004), not only is the state of Kansas short of health care providers but also has difficulty in recruiting them.

We are not the only military health Care facility facing lack of continuity of care for enrolled beneficiaries. Given the impact and the increasing number of cancer patients, the leadership within the United States Army Medical Command would find the results of this study useful. The MMCI model provides a measure to determine continuity of care. The moderate continuity findings, suggest a need to improve continuity of care. This will not only help the patient, but also reduce health care costs. Ultimately, the leadership, soldiers, family members, and retirees within the greater Fort Riley community will benefit from the results of this study.

This concept will help revise the management of continuity of care for all patients. It would further increase IACH’s reputation as providing efficient, effective, and quality health Care, while decreasing costs with a decrease in hospitalizations and emergency room visits.

References

Appendix A

(, 2000)

The star is the location of Irwin Army Community Hospital. The arrows designate referral facilities used by IACH. This allows the beneficiary population to receive the full spectrum of medical care.

Appendix B

(, 2000)

The star represents the location of Irwin Army Community Hospital. The majority of beneficiaries of the medical facility live within the oblong shape.

Appendix C

Comparisons of Continuity Scores

Components of Continuity Formulas and Score Comparison

UPC

COC

MCI

MMCI

1.00

0.52

1.00

1,1

0.50

0.00

0.05

0.10

1.00

0.68

1.00

2,1

0.67

0.33

0.35

0.51

1,1,1

0.33

0.00

0.03

0.04

1.00

0.76

1.00

3,1

0.75

0.50

0.51

0.67

2,2

0.50

0.33

0.51

0.67

2,1,1

0.50

0.17

0.27

0.36

1,1,1,1

0.25

0.00

0.02

0.03

1.00

0.80

1.00

4,1

0.80

0.60

0.61

0.76

3,2

0.60

0.40

0.61

0.76

3,1,1

0.60

0.30

0.41

0.51

2,2,1

0.40

0.20

0.41

0.51

2,1,1,1

0.60

0.10

0.22

0.28

1,1,1,1,1

0.20

0.00

0.02

0.03

Appendix D

Staging and Code Numbers for Cancer

Staging of Cancer

Table 1

Minimum, Maximum, Mean and Standard Deviations by Age, Years of Cancer, Number of Visits, Number of Providers, ER Visits, HD, TP, TV, MMCI And Evidence of Cancer

Variable

Min

Max

Age

52.58

15.08

Yrs cancer

7.99

6.41

Visits

176

23.04

21.66

Provider

9.52

8.86

Years seen

4.14

2.16

ER visits

230

3.87

21.24

Hospital

0.48

0.84

5.68

6.60

9.16

7.65

176

22.97

21.27

MMCI

0.59

0.21

Evid Canc.

1.14

0.34

Table 2

Frequency and Percent of Staging

Staging

Freq

Percent

12.7

50.0

8.5

15.3

3.4

0.8

0.0

9.3

Note. n=118.
Table 3

Minimum, Maximum, Mean And Standard Deviations By Age, Number Of Visits, Number Of Providers, Years Of Cancer, Years Seen, ER Visits, Number Of Hospitalizations, Number Of Day In Hospital, and MMCI

Variable

Min

Max

Age

53.39

9.66

Visits

24.30

17.74

Provider

9.70

6.25

Years Cancer

9.07

6.24

Years seen

4.09

1.93

ER visits

2.16

3.21

Hospital

0.48

0.85

Days

1.41

2.81

MMCI

0.12

0.58

0.22

Table 4

Frequency and Percent of Staging

Stage

Freq

Percent

11.4

61.4

4.5

22.7

Note. n=44

Table 5

Regression: B, Standard Error, Beta Weights, t, and Significance Levels for Variables and MMCI scores as the Criterion

Variables

Beta

Sig.

Out. Visits

.004

.002

.361

2.431

.020

ER Visit

-.019

.010

-.276

-1.861

.070

Hospital

.014

.037

.053

.367

.716

Table 6

Correlation for MMCI scores with Age, Race, Years with Cancer, and Years Seen

MMCI

Age

.09

Race

-.32*

Years cancer

-.40**

Years seen

-.03

Note. n=44, race (Caucasian=1, other=2), *p < .05, **p < 01.

DELIVERY OF CARE TO CANCER PATIENT

0 comments:

Post a Comment

DELIVERY OF CARE TO CANCER PATIENT

Next

Newer Post

Previous

Older Post

0 comments:

Post a Comment