The relationship between depressive symptomatology and high blood pressure in Hispanic elderly.
Subject: Depression, Mental (Psychological aspects)
Depression, Mental (Research)
Aged (Psychological aspects)
Aged (Health aspects)
Hispanic Americans (Psychological aspects)
Hispanic Americans (Health aspects)
Hypertension (Psychological aspects)
Hypertension (Research)
Authors: Rodriguez, Jose R.
Joglar, Priscille
Davila, Mariel G.
Pub Date: 06/22/2005
Publication: Name: American Journal of Health Studies Publisher: American Journal of Health Studies Audience: Professional Format: Newsletter Subject: Health Copyright: COPYRIGHT 2005 American Journal of Health Studies ISSN: 1090-0500
Issue: Date: Summer-Fall, 2005 Source Volume: 20 Source Issue: 3-4
Topic: Event Code: 310 Science & research
Product: Product Code: E123400 Hispanic Americans
Geographic: Geographic Scope: United States Geographic Code: 1USA United States
Accession Number: 157267622
Full Text: Abstract: The main purpose of this research was to examine the relationship between depressive symptomatology and the development of heart disease in Hispanic Puerto Rican elderly individuals and also to assess the validity of the Spanish translation of the Beck Depression Inventory (BDI-II) for this population. A probabilistic sample of 410 Hispanic men and women ages 65years and older was drawn from all of the residents in independent residential housing projects for elderly individuals in Puerto Rico. Our results suggest that depression alone is not associated with an increased risk for high blood pressure in Puerto Rican elderly living in independent residential housing projects.


A positive relationship between poor physical health and mental disorders, especially depression, has been established by many studies (Canino, Bird, Rubio-Stipec, & Bravo, 1997; Davidson, Jonas, Dixon, & Markovitz, 2000; Glassman & Shapiro, 1998; Jonas & Lando, 2000; Wells et al., 1989). This shows evidence of how mental and physical health are intimately related and complement each other. Many physical changes occur with the advancement of age that can lead to poorer health. These factors include decreases in mental and physical activity, negative cultural interpretations of aging, un-availability of medical resources and economic hardship factors. A pioneer research study was performed in Puerto Rico in 1997 using a probabilistic sample of older adults living in residential housing projects to assess their health status. The Rodriguez (1997) study found that 590/0 of the sample reported cardiovascular diseases and 60% of the elderly individuals report some form of depressive symptomatology. These findings suggest the need for in-depth exploration of the relationship between physical and mental health within elderly populations.

One of the physical conditions that can adversely affect elderly health is high blood pressure (HBP). According to the guidelines provided by the National Institutes of Health (NIH) and the National Heart, Lung and Blood Institute, hypertension is defined as systolic blood pressure of 140 mm Hg or greater; diastolic blood pressure of 90 mm Hg or greater; or taking antihypertensive medication (NIH, 1990). Hypertension is common in the elderly (NIH, 1990). Among Americans age 60 and older examined in the NHANES III, HBP was found in 60% of non-Hispanic whites and 61% of Mexican Americans (NIH, 1990). Systolic blood pressure is a better predictor of events like heart disease and mortality than is diastolic blood pressure, especially in older persons (NIH, 1990). Health statistics demonstrate that more than 50% of elderly individuals are at high risk to develop hypertension, or are suffering from it. Factors that affect hypertension include stressful life conditions, obesity, inadequate nutrition, smoking and lack of physical exercise; all of these factors promote poor physical health and makes elders prone to developing heart disease (Harris, 1991; Kart, Metress, & Metress, 1988).

Depression and heart disease are often companions. Numerous studies have demonstrated how one may affect the other, as well as multiple risk factors (Glassman & Shapiro, 1998; Mussleman, Evans, & Nemeroff, 1998; NIH, 1997). However, a clear and specific relationship between depression and HBP has not been established. Elderly individuals with depression and HBP have nearly twice the risk of having heart failure as people who had HBP alone (Abramson, Berger, Krumholz, & Viccarino, 2001).

The elderly population (65 and older) in Puerto Rico constitutes approximately 14% of the total population in the year 2000, and is expected to continue increasing at an alarming rate (U.S. Bureau of the Census, 2000). This fact will continue to have many implications for the island's public health care system and the prevalence of medical and psychological disorders. For example, the elderly report more mayor physical and mental health conditions than younger groups (Alsina & Rodriguez, 1995). Puerto Rican elderly are not only at the highest risk for committing suicide; they also have the highest rate of actual suicide prevalence rate on the island (Alsina & Rodriguez, 1995; Bravo, 2000; Rodriguez, 1997; Rodriguez & Collazo, 2003). Despite these alarming statistics, the most significant epidemiological research in this area that has been performed on the island is very limited and does not include any study of the prevalence, comorbidity and relationship between physical and psychological conditions in the elderly.

Moreover, there is extensive literature regarding depression as a risk factor for coronary heart disease, health problems, mortality and hypertension. (Barefoot & Schroll, 1996; Glassman & Shapiro, 1998; Jonas & Lando, 2000). However, there appears to be contradictory findings and results on the specific aspects of this issue. Some studies report a relationship between depression and hypertension (Aromaa et al., 1994; Frasure-Smith, Lesperance, & Talajic, 1995; Jonas, Franks, & Ingram, 1997; Reiff, Schwartz, & Northridge, 2001) while others report that they are not related (Paterniti, Alperovitch, Ducimetiere, Dealberto, Lepine, & Bisserbe, 1999; Raikkonen, Mathews, & Kuller, 2001; Shin, Poston, Kimball, St Jeor, Foreyt, 2001). An extensive review of the literature spanning 10 years reveal only one study that has statistically established a significant relationship between those two factors (Calvo, Dial, Ojeda, Ramal, & Aleman, 2001). Hence, this research attempts to elucidate the discrepancies found in the literature related to the role of depression and its influence on high blood pressure. It also provides valuable information concerning health disparities in the clinical and research arena.



Four hundred ten Hispanic men and women ages 65 years and older constituted our final number of participants. Participants were drawn probabilistically from all the residents in independent residential housing projects for elderly individuals in Puerto Rico. This sampling method is explained in more detail in a previous study (Rodriguez, 1997). Areas were divided based on a geographic location system used by the Puerto Rican Government's Office for Elderly Individuals Affairs (two regions). The specific sampling method for this study was random sampling by clusters. The formula to determine the number of participants by cluster was (Nh)(fh) = nh, where Nh is the number of independent residential housing projects for each region, fh is the standard deviation between sample and total population, and nh is the number of residents in each region. Using this formula we established that 316 participants from region I and 94 participants from region II were necessary to insure a representative sample of all the residents living in independent residential housing projects in Puerto Rico.

Table 1 presents descriptive demographic information of the sample. Women constituted 77% of the sample and 77% of the participants reside in urban regions. The most common medical conditions among the participants during the past six months were osteoarthritis (63%) and cardiovascular diseases (58%). Of the women in the study, 73% did not report gynecological conditions due to hysterectomy at early age and only 8% of the men reported prostate problems. Other conditions reported by this sample were stomach or intestinal disorders (30%), thyroid or endocrine disorders (24%), psychological disorders diagnosed by a mental health professional (16%), renal problems (15%), and pulmonary diseases (13%).


Demographic Data Questionnaire: A structured questionnaire was developed to collect relevant socio-medical demographic factors: age, gender, marital status, previous occupation, current income, education, alcohol, tobacco and illegal substance use, family history of high blood pressure and coronary heart disease, history of previous depression diagnoses, suicide attempt during the last 10 years, other diagnosed mental and physical conditions (including hypertension) during the last 10 years, and duration of time living in the independent residential housing project.

Beck Depressive Inventory-II: The Beck Depressive Inventory (BDI-II) has been used extensively in many research projects in the United States and Puerto Rico; however, its Spanish translation has never been appropriately validated with the Hispanic elderly. Nevertheless, it has been used within the general Puerto Rican population in specific research search projects that explore its psychometric properties (Bravo, 2000; Lugo & Rodriguez, 1997). The 21-item instrument was administered to the Hispanic elderly to assess its internal consistency and construct validity on this population.


After receiving permission from the administrators of independent residential housing projects in both regions, researchers made personal contact verbally and in writing to explain the nature and aims of the project, and to assuage any doubts before seeking permission from the elderly residents. Once the elderly individuals were invited to participate in the study and agreed to sign a consent form, trained doctoral graduate students in psychology (TDGSP) administered the BDI-II and demographic data questionnaire. The instruments were administered orally to each participant within a confidential.

In order to obtain a quantitative blood pressure (BP) measure, graduate students used an automatic digital commercial sphygmomanometer placed on the left arm of participants who were seated. The BP measure was taken three times within a 20-30 minute interval. The first measurement was taken after the consent form was signed by the elderly individual (baseline). The second measurement was taken prior to beginning the BDI-II (pre-test). The third measurement was taken after the administration of the scale (post-test). Median BP measures were selected as the measure of record for the statistical and correlational analyses. According to the guidelines provided by the NIH and the National Heart, Lung, and Blood Institute, participants were classified as hypertensive if their median score for three measurements of systolic blood pressure was 140 mm Hg or greater or the median score for diastolic blood pressure was 90 mm Hg or greater (NIH, 1997). Proper referrals were reported to the administrators of the residential housings, as necessary.


The present study consisted of two phases. The first phase was designed in order to determine the reliability and validity of the BDI-II. The second phase had an ex post facto design that explored the correlation between depressive symptomatology and hypertension. To our knowledge, there are no studies in Puerto Rico that have assessed the relationship between the rate of depressive symptomatology and hypertension in elderly individuals.


SPSS (release 12.0 for Windows) was used for all data analyses. Specifically, BD-II psychometric properties were analyzed via Chronbach's alpha and Spearman-Brown analysis for internal consistency. A confirmatory factor analysis was also performed to on the BDI-II. Additional analyses included descriptive analyses of demographic variables, correlations between nominal variables using cross-tabulation (Chi-square), differences between using Student's t test and analysis of variance (ANOVA), correlation between the main variables using Pearson's product moment (r), and a hierarchical regression analysis for specific variables ([R.sup.2]). The alpha level for the analyses was p d * 0.05.



Reliability and validity of the Spanish translation of the BDI. A Chronbach's alpha of .89 and a Spearman-Brown value of .85 indicated a high internal consistency of the BDI-II as applied to the assessment of depressive symptomatology in Hispanic Puerto Rican elderly. In addition, factor analysis was performed in order to explore the construct validity of the BDI-II. Specifically, analysis utilized the principal component factoring extraction method with orthogonal rotation (Varimax). The number of factors extracted was four and these factors accounted for 52% of the total variance of the instrument. This finding was consistent with the results obtained in previous studies regarding the psychometric properties of the BDI-II (Bravo, 2000; Lugo & Rodriguez, 1997).

Relationship between depressive symptomatology and high blood pressure. Analyses revealed a negative, low, and non-significant correlation between depression (assessed by the score on the BDI-II) and median systolic pressure (r = -.025, p = .61). Also, based on the BD-II results, and controlling for the number of anti-hypertensive medicines taken, there was a positive, low, and non-significant correlation between depressive symptomatology and diastolic pressure (r = .032, p = .52). These findings are consistent with the results obtained by previous research in which they established that depression was not independently associated with an increased risk for high blood pressure (Paterniti et al., 1999; Raikkonen et al., 2001; Shin et al., 2001). Interestingly, controlling for the number of anti-hypertensive medicines taken, a negative, moderately low, and significant correlation was found between BDI-II scores and reported depression (a qualitative measure/self-report; r = -.50, p = .001).

A one-way repeated measure ANOVA was calculated comparing participants' blood pressure measurements at three different times: baseline, pretest, and post-test. A significant effect was found for systolic pressure [F (1,409) = 20.48, p = .001], but not for diastolic pressure [F (1,409) = .061, p = .80]. Follow-up protected t tests for systolic pressure revealed that scores decreased significantly from baseline to pre-test (t = 5.03, p = .001) and from baseline to post-test (t = 4.53, p = .001). No significant effect was found between pre-test and post-test measures (t = -.41, p = .68). The instrumentation effect of the BDI-II was meaningfulness between administrations. This finding suggests that, as an instrument, the BDI-II had no influence over the pre-test and post-test blood pressure measurements.


Researchers obtained a mean BDI-II score of 7.90 points (SD = 7.63), a mean systolic pressure of 141.48 (SD = 20.13), and a mean diastolic pressure of 78.96 (SD = 13.42) for the total sample. Based on these scores, participants in the sample seem to be hypertensive, but not depressed.

According to the literature reviewed, variables such as depression and high blood pressure may occur at different rates in women and men. This difference also may be related to their area of residence. Study results revealed that no significant differences between geographic regions (Region I and Region II) in terms of BDI-II scores (t = -.86, p = .39), systolic pressure (t = 1.51, p = .13), and diastolic pressure (t = .58, p = .56). Also, there were no significant differences by gender in relation to BDI-II scores (t = -.10, p = .93), systolic pressure (t = 1.74, p = .08), and diastolic pressure (t = -.27, p = .79).


Hierarchical regression analysis was conducted in order to explore which variables are good predictors for BDI-II scores. Results indicated that the higher the number of medical conditions an individual suffers, the higher the score he or she will obtain on the BDI-II. This result was independent of age, educational level (associated with a better repertoire of coping skills), systolic and diastolic measurements ([sup.~] [R.sup.2] = .16, p < .05; see Table 2).


There is compelling evidence that elderly Hispanic Puerto Ricans are affected by mental health variables that may also affect their mortality rates (Barefoot & Schroll, 1996; Mussleman, et al., 1998; Wells et al., 1989). Previous research performed in Puerto Rico has found that 62% of the elderly sample was suffering from depressive symptomatology (Rodriguez, 1997). Moreover, findings of a retrospective epidemiological study indicated that the suicide rate for those who are 85 years old or older was 12.58 while it was only 7.40 for those who are 64 years old or younger (Alsina & Rodriguez, 1995). These findings have important implications since approximately 15% of the total population in the year 2000 were individuals 65 years and older and this population utilizes the highest amount of health services (Alegria et al., 2001). The present study was conducted in order to explore the physical problems of Hispanic Puerto Rican elderly individuals.

This research study principally examined the extent to which depression symptomatology was associated with high blood pressure measurements. The existing literature on this relationship seemed contradictory. Some studies demonstrated a possible association between the two variables (Aromaa et al., 1994; Frasure-Smith et al., 1995; Jonas et al., 1997; Reiff et al., 2001) and others demonstrating evidence to the contrary (Paterniti et al., 1999; Raikkonen et al., 2001; Shin et al., 2001;). Our findings suggest that there is no statistically significant correlation between the variables in this population.

Although the BDI-II proved to be a reliable and valid instrument to assess depressive symptomatology in this population, one must be cautious in this assertion. This study provides evidence regarding the expected influence of two possible subtypes of depressive symptoms on hypertension per se. Clinical literature shows that depression could have two different manifestations of symptoms, the negative affect melancholic type of symptoms that involve hypersomnia or eating more than usual, and the depression involving manic states characterized by overeating and insomnia (NIH, 1997). As shown in several studies, people suffering from depression with negative affect (melancholic) that involves a decrease in desire to participate in activity or sadness have a tendency towards poor physical and social functioning that may lead to poor cardiovascular health (Jonas & Lando, 2000; NIH, 1997; Rodriguez, 1997, Wells et al., 1989). Mainly, the participants in our study did not manifest this type of melancholic depressive symptoms, or negative affect. The BDI-II items in general were designed, as identified by factor analysis, to gain information primarily about negative affect. This explains why the hypertensive participants in this study reported feeling depressed in spite of low BD-II scores (5 to 9 points), indicating absence of depressive symptomatology. This fact may explain why results showed a significant correlation between reported or perceived depression, a more qualitative measure, and BDI-II scores. But as previously discussed, it could be argued that participants' depression was better accounted by positive affect instead of negative. However, the essential interpretation and importance of the BDI-II reliability and validity for this population remains unchanged.

According to study results, a good predictor of high scores on the BDI-II would be the number of medical conditions an individual suffers independently of specific factors such as blood pressure measurements, age, and educational level. The latter are usually associated with a better repertoire of coping skills and other cognitive strategies to manage stressful situations. Thus, scores on the BDI-II will be more susceptible to change as the amount of physical complaints increases. The intensity or magnitudes of those complaints were not addressed in this study; however it would be wise to conduct future research in this area. Finally, other aspects consistently associated with depressive symptomatology and high blood pressure must await future studies (e.g.,, exercise, self-reported depression).

Taking into account the probabilistic nature of our sample and a response rate of a 100%, we can reaffirm the statistical power of this investigation. This high response rate might be due to the type of population studied. The elderly in Puerto Rico live in government housings where all the residents are of the same age and eager to share and talk to outside visitors. Even so, despite careful measures there are several limitations to the current study. One potential limitation to using the BDI-II to measure symptoms of depression is that it mainly assesses negative affect. As previously discussed, positive affect can be a part of depression as well. However, the BDI-II is one of the most widely clinically used and well-validated self-report measure of depressive symptomatology in the United States and now in Puerto Rico. Future investigations should consider the use of additional, culturally sensitive and more comprising instruments that consider the specific population studied and achieve a more rigorous assessment. Researchers in the area of test construction need to actively advocate regarding this issue. It can be more suitable for methodological purposes to classify participants according to the subtype of depression that they are experiencing prior to the study. There needs to be more research done on depression, the continuum of manic to melancholic symptoms related to the condition, and the impact that differences in depressive symptoms could have on the development of cardiovascular disease or hypertension.

This research was funded by the National Institute of Health (NIH), Grant # 1 R24MD00152-0. It has been accepted as a poster section in the 132nd Annual Meeting of the American Public Health Association (Session 3037.0) in Washington, DC.


Abramson, J., Berger, A., Krumholz, H., & Viccarino, V. (2001). Depression and risk of heart failure among older persons with isolated systolic hypertension. Archives of Internal Medicine, 161(14), 1725-1730.

Alegria, M., McGuire, T., Vera, M., Canino, G., Albizu, C., Marin, H., et al. (2001). Does managed mental health care reallocate resources to those with greater need for service? Journal of Behavioral Service Resource, 28, 439-455.

Alsina, S., & Rodriguez, J. (1995). El fenomeno del suicidio en el anciano puertorriqueno. [The suicide phenomenon in the Puerto Rican elder]. Revista de Psicologia Latino Americana, 27(2), 263-282.

Aromaa, A., Raitasalo, R., Reunanen, A., Impivaara, O., Heliovaara, M., Knekt, P., et al. (1994). Depression and cardiovascular diseases. Acta Psychiatrica Scandinavica (Suppl 377), 77-82.

Barefoot, J., & Schroll, M. (1996). Symptoms of depression, acute myocardial infraction, and total mortality in a community sample. Circulation, 93(10), 1976-1980.

Bravo, L.M. (2000). Factores que intervienen en los niveles de sintomatologia depresiva de los ancianos/ as en las egidas de Puerto Rico. [Factors that influence the depressive symptomatology levels of the Puerto Rican elderly living at independent housings]. Unpublished doctoral dissertation, Carlos Albizu Universtiy, Puerto Rico.

Calvo, F., Diaz, D., Ojeda, B., Ramal, J., & Aleman, S. (2001). Diferencias en ansiedad, depresion, estres y apoyo social entre sujetos hipertensos y normotensos. [Differences in anxiety, depression, stress, and social support between hypertensive and non-hypertensive individuals]. Ansiedad y Estres, 7(2-3), 203213.

Canino, G., Bird, H., Rubio-Stipec, M., & Bravo, M. (1997). The epidemiology of mental disorders in the adult population of Puerto Rico. Puerto Rico Health Science Journal 16(2), 117-124.

Davidson, K., Jonas, B., Dixon, K., & Markovitz, J. (2000). Do depression symptoms predict early hypertension incidence in young adults in the cardiac study? Archives Internal Medicine, 160, 1495-1500.

Frasure-Smith, N., Lesperance, F., & Talajic, M. (1995). The impact of negative emotions on prognosis following myocardial infraction: Is it more than depression? Health Psychology, 14(5), 388-398.

Glassman, A. & Shapiro, P. (1998). Depression and the course of coronary artery disease. American Journal of Psychiatry, 155(1), 4-11.

Harris, D. (1991). Sociology of aging. New York: Harper & Row.

Jonas, B., Franks, P., & Ingram, D. (1997). Are symptoms of anxiety and depression risk factors for hypertension? Arch Fam Med, 6, 43-49.

Jonas, B., & Lando, J. (2000). Negative affect as a prospective risk factor for hypertension. Psychosomatic Medicine, 62, 188-196.

Kart, C., Metress, E., & Metress, S. (1988). Aging, health and society. Boston, MA: Jones & Barlett.

Lugo, Y., & Rodriguez, J. (1997). Estudio piloto sobre la ideacion suicida en ancianos puertorriquenos en un hospital de salud mental. [A pilot study about suicide ideation in a inpatient Puerto Rican elderly sample]. Boletin de la Asociacion Medica de Puerto Rico, 89(10-12), 167-173.

Mussleman, D., Evans, D., & Nemeroff, C. (1998). The relationship of depression to cardiovascular disease. Archives of General Psychiatry, 55(7), 580.

National Institutes of Health (NIH). (1997). The sixth report of the joint national committee on prevention, detection, evaluation, and treatment of high blood pressure (Publication No. 98-4080). Washington, DC: Government Printing Office.

Paterniti, S., Alperovitch, A., Ducimetiere, P., Dealberto, M., Lepine, J., & Bisserbe, J. (1999). Anxiety but not depression is associated with elevated blood pressure in a community group of French elderly. Psychosomatic Medicine, 61, 77-83.

Raikkonen, K., Mathews, K., & Kuller, L. (2001). Trajectory of psychological risk and incident hypertension in middle aged women. Hypertension, 38(4), 798-802.

Reiff, M., Schwartz, S., & Northridge, M. (2001). Relationship of depressive symptoms to hypertension in a household survey in Harlem. Psychosomatic Medicine, 63, 711-721.

Rodriguez, J. (1997). Variables sociodemograficas, apoyo Social, depresion y estado de Salud de ancianos/ as de las egidas de Puerto Rico: Un estudio descriptivo. [Socio-demographic variables, social support, depression, and health status in a Puerto Rican elderly sample living at independent housings: A descriptive study]. Oficina Para Asuntos de la Vejez de la Gobernadora de Puerto Rico, San Juan, Puerto Rico.

Rodriguez, J. & Collazo, L. (2003). Estudio epidemiologico retrospectivo sobre el fenomeno del suicidio en los ancianos puertorriquenos: Datos de dos decadas. [An epidemiological retrospective study about the suicide phenomenon in the Puerto Rican elderly]. Thompson Psicologia, 1, 77-87.

Shin, E., Poston, W., Kimball, K., St Jeor, S., & Foreyt, J. (2001). Blood pressure and symptoms of depression and anxiety: A prospective study. American Journal Hypertension, 14(7 Pt 1), 660-664.

U.S. Bureau of the Census. (2000). The 2000 U.S. census report. Retrieved February 20, 2003, from

Wells, K., Stewart, A., Hays, R., Burnam, M., Rogers, W., & Daniels, M., et al. (1989). The functioning and well being of depressed patients. Journal American Medical Association, 262(7), 914-919.


Responsibility IV - Evaluating Effectiveness of Health Education Programs

Competency A--Develop Plans to Assess Achievement of Program Objectives

Jose Rodriguez, MD, MPH, PhD, Priseille Joglar, PsyD, and Mariel G. Davila, PhD are affiliated with Carlos Albizu University. Address all correspondence to Jose Rodriguez, MD, MPH, PhD, Carlos Albizu University, PO Box 9023711, San Juan, Puerto Rico 00902-3711. PHONE: 787-725-6500 x125; E-MAIL:
Table 1. Socio-medical Demographics.

             Variable                 N     %       M        SD

Men                                   94    23
Women                                316    77
Geographic Region
  Region                             131   677
  Region II                           94    23
Marital Status
  Married                             40    10
  Widowed                            191    47
  Divorced                            98    24
  Never Married                       80    19
Work Status
  Not Working                        398    97
  Working                              8     2
  Volunteering                         4     1
  Catholic                           257    63
  Protestants                        130    32
  No Preference                       23     5

Age                                                 75.6       7.3
Reported Income (annual)                         5,057.1   2,769.9
Education (years)                                    8.3       4.5
Time living in Housing (years)                       7.6       6.2
Number of Medical Conditions (life
  time prevalence)                                   3.3       2.1

Table 2. Summary of Hierarchical Regression Analysis for
Variables Predicting BDI-II Scores (N = 410)

                  Variables                B     SE    [B.sup.~]
Step 1
         Systolic Pressure               -.002   .02     -.06
         Diastolic Pressure               .003   .03      .06
Step 2
         Number of Medical Conditions    1.37    .16      .38 *
         Systolic Pressure               -.003   .02     -.07
         Diastolic Pressure               .004   .03      .07
Step 3
         Number of Medical Conditions    1.33    .16      .37 *
         Age                             -.008   .05     -.07
         Education                      -0.12    .08     -.07
         Systolic Pressure               -.002   .02     -.06
         Diastolic Pressure               .003   .03      .06

Note: [R.sup.2] = .003 for Step 1; [sup.~] [R.sup.2] = .15 for
Step 2; [sup.~] [R.sup.2] = .16 for Step 3 (Ps < .05).

B = regression coefficient; SE B = standard error of the
regression coefficient; ~ = standardized regression

* p < .05.
Gale Copyright: Copyright 2005 Gale, Cengage Learning. All rights reserved.