J Epidemiol Community Health: first published as 10.1136/jech.52.3.186 on 1 March 1998. Downloaded from 186 J Epidemiol Community Health 1998;52:186–190 John Henryism and among Nigerian civil servants

Nina Markovic, Clareann H Bunker, Flora A M Ukoli, Lewis H Kuller

Abstract African-Americans of low socioeconomic sta- Study objective—Among urban Nigerian tus or of low educational attainment, those civil servants, higher socioeconomic sta- with high John Henryism scores show in- tus is related to increased blood pressure. creased blood pressure or increased prevalence In the United States, the relation between of .9–12 This finding has also been increased blood pressure and low socio- confirmed in a white Dutch population.13 More economic status or low level of education recently, high John Henryism has also been has been found to be potentiated by high found to potentiate an association between eVort active coping (John Henryism) high job status and blood pressure among among African-Americans. Thus, the po- women and African-Americans employed in tentiating eVect of high eVort active the United States.14 However, the combination coping as measured by the John Henryism of high John Henryism and low education was Active Coping Scale, on socioeconomic not associated with increased blood pressure in status, as measured by job grade, was a young cohort (18–30 year olds).15 considered in relation to blood pressure in Among civil servants in Nigeria, we have a Nigerian civil servant population. observed a positive association between blood Design—The influence of John Henryism pressure and socioeconomic status, with in- on the association between educational creased blood pressure measurements among level or socioeconomic status and in- those of higher staV status.16 During our most creased blood pressure was examined recent comprehensive blood pressure survey of during a comprehensive blood pressure civil servants in Benin City, Nigeria, we used survey. John Henryism refers to a strong the John Henryism Active Coping Scale17 to behavioural predisposition to actively examine its application in this non-Westernised cope with psychosocial environmental population. We wanted to replicate the analyses stressors. of James et al,9–12 that is, to test the hypothesis Setting—Benin City, Nigeria. that those with lesser material coping resources Participants—Nigerian civil servant sam- (for example, lower educational attainment, ple of 658 adults, aged 20 to 65 years. lower job grade) who scored high on the John Main results—Among those with high Henryism Active Coping Scale, would have John Henryism scores of upper socio- increased blood pressure. Alternatively, we economic status, whether measured by sought to examine whether or not this measure education level or job grade, there was a of active coping further explained our previous http://jech.bmj.com/ trend toward higher systolic and diastolic finding of increased blood pressure among blood pressures, adjusted for age and body those of higher staV status. mass index, in men and women, though not statistically significant. Department of —This trend is consistent Methods , Conclusions with recent findings of increased blood We recruited 84% of the civil servants at three Graduate School of ministries of Edo State in Benin City, Nigeria, Public Health, pressure among women and African- University of Americans with high John Henryism and for participation in a comprehensive blood on September 27, 2021 by guest. Protected copyright. Pittsburgh, Pittsburgh, high status jobs. pressure survey. This population has been pre- USA viously described18 and is characterised as a N Markovic (J Epidemiol Community Health 1998;52:186–190) lean population with a low protein and fat C H Bunker intake, moderate sodium intake, and a very low L H Kuller rate of smoking. These analyses are restricted Department of Increased blood pressure has been seen to be to the 658 (81.3%) of the 809 civil servants, Community Health, associated with a number of physical, socio- between the ages of 20 and 65 years, who com- University of Benin demographic, and psychological factors.1–5 Nu- pleted the John Henryism Active Coping Scale. Teaching Hospital, merous epidemiological studies have found Administration of the John Henryism Active Benin City, Edo State, socioeconomic status to be inversely related to Coping Scale was begun after the start of study Nigeria F A M Ukoli increased blood pressure among populations in data collection, thus the first 151 participants European countries and the United States.6–8 did not complete the John Henryism Active Correspondence to: Among African-Americans residing in the Coping Scale. Each individual was seen at Nina Markovic, Department United States, the relation is thought to be three morning appointments scheduled over of Epidemiology, Graduate School of Public Health, influenced, in part, by a coping strategy called one week at the study site located in each of the University of Pittsburgh, 130 “John Henryism”. John Henryism has been respective ministries. At each of the three visits, DeSoto Street, Pittsburgh, defined as a strong behavioural predisposition three blood pressure determinations were PA 15261, USA. to cope in an active, determined, and hard- made by Nigerian study team members, certi- 19 Accepted for publication working manner with the stressors of everyday fied by a standard protocol. First and fifth 22 April 1997 life.9 The influence is such that among KorotoV sounds were recorded as systolic and J Epidemiol Community Health: first published as 10.1136/jech.52.3.186 on 1 March 1998. Downloaded from Nigerians, John Henryism, and blood pressure 187

Table 1 Mean systolic (SBP) and diastolic blood pressure (DBP)* for the four groups were constructed for assessing educa- education-John Henryism (JH) groups by sex, Benin City Nigeria Civil Servants 1992 tion and John Henryism, based on the median splits of the John Henryism Active Coping Scale Men Women and educational attainment of the sample.9 For Number SBP DBP Number SBP DBP the second presented analysis, four contrast groups were constructed based upon the High education: High JH 74 122.4 78.3 47 115.8 73.7 median splits of the John Henryism Active Low JH 63 119.2 77.7 32 110.1 72.7 Coping Scale score and job grade status (Junior Low education: or Senior)16 for the analysis of sta status and High JH 140 121.6 77.5 92 113.6 71.7 V Low JH 126 118.1 75.4 84 111.3 72.0 John Henryism. Entire sample 403 120.3 77.0 255 112.8 72.3 ANOVA p values: Education 0.76 0.74 0.71 0.65 Results John Henryism 0.10 0.26 0.04 0.82 The John Henryism Active Coping Scale Education × JH 0.73 0.36 0.26 0.46 scores among these Nigerian civil servants Post hoc ANOVA: Between high JH and high education ranged from 22 to 60, with a highly skewed versus all others combined p=0.75 p=0.88 p=0.20 p=0.39 distribution to the higher end of the scale. The median John Henryism Active Coping Scale * Mean systolic and diastolic blood pressures are an average of six readings across three visits, adjusted for age and body mass index. score in this population was 57, with 16.8 per cent of the respondents scoring 60. The John diastolic blood pressure respectively. Mean Henryism score mean and standard deviation blood pressures calculated as the average of the were 55.9 and 4.14. The median for education second and third determinations made at each was completion of secondary level education, of the three study visits were used in all so that those who had completed technical presented analyses. school beyond secondary school or university Demographic data were collected at the first were grouped as “high” education. visit, and included recording of level of educa- John Henryism Active Coping Scale scores tion attained, job grade, marital status, ethnic did not diVer significantly between men and tribe, height, and weight (in light clothing women (mean scores of 55.8 and 56.1 respec- without shoes). Education was categorised tively), nor were there significant diVerences from no education, completion of primary between women and men in the proportion education, completion of secondary education, classified as “high John Henryism” (per cent completion of technical education, to comple- classified as high John Henryism among men tion of a university degree. Nigerian civil serv- was 78.7 and among women 80.4). No signifi- ant job grades are ranked from 1 to 16, such cant correlations were noted between John that grades 1 to 6 are considered Junior staV Henryism scores and age (r=0.0002, p>0.05), (unskilled workers, semi-skilled workers, and body mass index (r=0.038, p>0.05), job grade entry level administrators), while those of level (r=0.011, p>0.05), or level of formal edu- grades 7 to 16 are considered Senior staV cation (r=0.028, p>0.05). No significant diVer- (highly skilled workers and administrators). ences were identified among marital status cat- Marital status was recorded as never married, egories or among the ethnic tribes. married, divorced, and widowed. Ethnic tribal Mean systolic and diastolic blood pressures

aYliation was self reported. Body mass index were positively associated with levels of educa- http://jech.bmj.com/ was calculated as kg/m2. tion (r=0.03, r=0.06 respectively) and staV Before administration the John Henryism level (r=0.18, r=0.19 respectively), age Active Coping Scale was translated into pidgin (r=0.04, r=0.28 respectively) and body mass English, which is the common language for this index (r=0.16, r=0.19, respectively). These population. Back translation by another Nige- correlations were all significant at p<0.001 rian study team member who was unfamiliar except for level of education. In univariate with the John Henryism Active Coping Scale analyses of known risk factors, age, body mass

was acceptable, and the John Henryism Active index, waist to hip ratio, number of years urban on September 27, 2021 by guest. Protected copyright. Coping Scale was completed by interview to living, number of children, and blood glucose avoid potential bias between literate and illiter- concentration were significantly associated ate respondents. with blood pressure among the entire popula- Data were analysed by use of the statistical tion. However, only age, body mass index, and package SPSS-PC+ (SPSS-PC+ Inc, Chicago, job grade remained significant (p<0.05) in the IL). Correlations between John Henryism and linear regression models for blood pressures. age, job grade level, and education level were Because job grade was used in the John Henry- conducted. The ÷2 test for diVerences in John ism construct, only age and body mass index Henryism between men and women, senior were adjusted for in analysis of diVerences staV and junior staV, marital status, and ethnic between the John Henryism contrast groups. tribes was also conducted. Contrast groups Men demonstrated significantly higher systolic were compared using analysis of variance and diastolic blood pressures, and sex specific (ANOVA). The mean systolic and diastolic data are presented for analysis of diVerences blood pressures were adjusted for age and body between John Henryism contrast groups (mean mass index in the analysis, and sex specific data systolic and diastolic blood pressures among were analysed. men was 120/77 mm Hg; and among women For analyses of blood pressure outcomes, 113/72 mm Hg, p<0.01). contrast groups were established according to Among the men, adjusted mean systolic and the methodologies of James and coworkers.9–11 13 diastolic blood pressures were higher in Senior In the first presented analysis, four contrast staV than in Junior staV (123.6 mm Hg/79.2 J Epidemiol Community Health: first published as 10.1136/jech.52.3.186 on 1 March 1998. Downloaded from 188 Markovic, Bunker, Ukoli, et al

Table 2 Mean systolic (SBP) and diastolic blood pressure (DBP)* for the staV Discussion status-John Henryism (JH) groups by sex, Benin City Nigeria Civil Servants 1992 In accordance with our previous study of Nige- rian civil servants, we noted higher systolic and Men Women diastolic blood pressures among the upper Number SBP DBP Number SBP DBP socioeconomic status group, as categorised by civil servant job grades. This is in contrast with Senior staV: High JH 97 125.8 79.7 62 115.1 73.3 findings in US and European populations Low JH 79 119.0 77.2 48 111.5 73.2 where there is an inverse relation between Junior staV: High JH 117 118.6 76.2 77 113.8 71.7 blood pressure and socioeconomic status. In Low JH 110 118.0 75.4 68 110.6 71.5 this Nigerian population, we observed a Entire sample 403 120.3 77.0 255 112.8 72.3 non-significant trend toward increased blood ANOVA p values: StaV status 0.26 0.23 0.19 0.76 pressures with higher education level, and with John Henryism 0.11 0.27 0.04 0.79 higher staV status, among those with higher StaV status × JH 0.26 0.84 0.59 0.55 John Henryism. This finding is consistent with Post hoc ANOVA: Between high JH and senior staV the hypothesis that high eVort coping among versus all others combined p=0.03 p=0.15 p=0.96 p=0.69 those of higher job status may result in blood pressure increases among Nigerian civil serv- * Mean systolic and diastolic blood pressures are an average of six readings across three visits, adjusted for age and body mass index. ants. In post hoc analyses, significantly higher systolic blood pressure was observed among mm Hg versus 117.8 mm Hg/75.5 mm Hg, Senior male staV with high John Henryism respectively, p<0.05). Significant diVerences in compared with all other men. Those with lesser adjusted blood pressures were not observed coping resources, such as lower educational between women of Senior and Junior staV sta- attainment, or lower job grade, and high John tus, nor between high and low education Henryism did not demonstrate significantly contrast groups in either sex. Table 1 shows the increased systolic or diastolic blood pressure. comparison of adjusted mean systolic and In earlier studies of US black populations,9–11 diastolic blood pressure in high/low John Hen- it was the lower socioeconomic status stratum ryism groups across high/low education that had the higher blood pressure, and blood groups. DiVerences across the four John pressure was even higher among those of both Henryism education groups in men were small lower socioeconomic status and high John and not significant. Among women, systolic Henryism. It was hypothesised that blood blood pressure was higher in both high John pressure was aVected by stresses resulting from high coping eVorts among those with very lim- Henryism groups (p=0.04), and was highest in ited resources for coping. However, the find- the high education group. ings in this study are consistent with a more When examined in relation to staV status recent study14 that found higher blood pressure (table 2) high John Henryism had no influence among women and among black men of higher on blood pressure among the men of Junior occupational status and high John Henryism. staV status. However, among the men of Senior These groups should have relatively high staV status, higher blood pressure was observed coping resources suggesting that the eVects of among the high John Henryism group, while John Henryism are not dependent on limited

the low John Henryism group had mean blood coping resources. Light and colleagues found http://jech.bmj.com/ pressures similar to the men of Junior staV sta- high blood pressure among women and among tus. Among women, systolic blood pressure black men of higher occupational status and was higher in both high John Henryism groups high John Henryism, and suggested that the (p=0.04) and was highest among the Senior of breaking down traditional barriers for staV. Of note, the Senior staV in the high John occupational advancement among women and Henryism group had the highest blood pres- African-Americans in the United States may sure among both men and women. Although interact with John Henryism and the associa- the formal ANOVA test for interaction between tion with increased blood pressure on September 27, 2021 by guest. Protected copyright. 14 staV and John Henryism status was not signifi- measurements. This leads to an alternative cant, post hoc analyses of adjusted blood pres- hypothesis that John Henryism interacts with sure was significantly higher in the high John other factors predisposing to higher blood Henryism/Senior male staV compared with the pressure, whether the stresses of low socio- economic status or the stress of breaking down other three contrast groups combined (high traditional barriers to occupational advance- John Henryism/Senior male staV 125.8 mm ment or other unidentified stresses and risk Hg/79.7 mm Hg versus 118.5 mm Hg/76.2 factors. mm Hg all other men, p=0.03 systolic, p=0.15 We speculate that several factors may have diastolic). Among women, those catorgorised influenced the significance of our findings. as high John Henryism/Senior staV status did Firstly, although the literal translation of the not have significantly higher systolic or diasto- John Henryism Active Coping Scale was lic blood pressures compared with the other acceptable, the contextual interpretation may three contrast groups combined (high John not have been adequate in identifying those Henryism/Senior female staV 115.1 mm Hg/ with an excessive determination or drive to 73.3 mm Hg versus 112.1 mm Hg/72.0 mm actively cope with behavioural stressors in the Hg all other women, p=0.96 systolic, p=0.69 environment. In this economically developing diastolic). StaV status was no longer signifi- country where many lack material coping cantly related to blood pressure after the inclu- resources, Nigerians struggle against enormous sion of John Henryism in the analyses. odds daily to accomplish what may be consid- J Epidemiol Community Health: first published as 10.1136/jech.52.3.186 on 1 March 1998. Downloaded from Nigerians, John Henryism, and blood pressure 189

ered simple everyday task, such as obtaining water. Thus, active coping with stressors in the KEY POINTS environment—that is, John Henryism—is a + Previously reported findings of increased way of routine life for all levels of socioeco- blood pressure among Nigerian senior nomic status in Nigeria. staV are not wholly explained by body Additionally, this Nigerian civil servant mass index and age. population is considered reflective of a society + In west African urban civil servants of in transition to a more “westernised” lifestyle. upper socioeconomic status, with high It is unclear whether a cultural tradition of scores on the John Henry Active Coping highly active coping or if high levels of coping Scale, there was a trend towards higher are required to accompany the social change blood pressures adjusted for age and leading to westernisation is reflected in our body mass index. finding of high John Henryism among the + This trend is consistent with recent find- entire spectrum of civil servants. These analy- ings of increased blood pressure among ses are based upon cross sectional data, and it women and African-Americans with high is unclear where in the continuum of the John Henryism and high status jobs. changing gradient between socioeconomic sta- tus and cardiovascular disease risk these Nige- rian civil servants occupy. It may be that this sors, such as Dressler’s “lifestyle incongruity”20 population in early westernisation reflect a are being explored among this population. midpoint in the continuum, and the population The protocol used to measure blood pres- is experiencing high levels of coping during this sure in this survey is not considered to have period of transition. As the association between contributed to the observed blood pressure socioeconomic status and cardiovascular dis- diVerences among the staV status, educational ease risk becomes indirect, measures of John level, or John Henryism contrast groups. The Henryism will reflect personal intrinsic charac- possibility of blood pressure measurement as teristics rather than reflect a response to social being a novel experience, reflected in an changes. increased blood pressure (a “white coat” Secondly, in our study sample of civil response) has been considered. Previous analy- servants, the low end of socioeconomic status ses of blood pressure determinations con- and educational attaintment levels in Nigerian ducted among this population did not show society—that is, the unemployed and village any significant variability among the staV status farmers— were not surveyed. Without those of groupings nor sex in the repeated blood the lowest end of material coping resources, we pressure measurements21 that would aVect may have lacked the contrast necessary to these results. Indeed, we do not find evidence observe an association. This is similar to James that repeated blood pressure determination and coworkers findings in American whites.10 variability is greater among these urban dwell- Thirdly, by administering the John Henry- ing Nigerian civil servants compared with other ism Active Coping Scale through interview, populations in either developed or developing scores may have been inflated as a result of the countries. social saliency of the more positive responses. In conclusion, we noted a consistent trend of This is reflected in the large percentage of the higher blood pressures for those of high John http://jech.bmj.com/ sample that scored 60, and the high median Henryism and high education or staV status in score of 57. Studies using the John Henryism this Nigerian civil servant population. Further Active Coping Scale conducted in the United research of the John Henryism Active Coping States and the Netherlands also report highly Scale among all social strata in economically skewed distributions with median scores of developing countries is suggested to determine 54.91013 its application in such populations. Finally, it may be that those classified as low on September 27, 2021 by guest. Protected copyright. socioeconomic status in this civil servant We gratefully acknowledge the cooperation of the Edo State of population represent those who have more Nigeria Ministries of Education, Information, and Agriculture. The authors also thank the manuscript referees for their adequate coping resources compared with comments, which enhanced the quality of this manuscript. their similarly educated peers who face very high unemployment rates. Thus, a secure civil Funding: this research was supported by Research Grant servant job among those of lower socioeco- HL44413 (US National Institutes of Health) and Training Grant HL07011 (USNational Institutes of Health). nomic status may attenuate stressors, resulting Conflicts of interest: none. in lowered sympathetic arousal and a lower risk for hypertension, as demonstrated by the lower 1 Stamler J, Rhomberg P, Schoenberger JA, et al. Multivariate mean systolic and diastolic blood pressures in analysis of the relationship of seven variables to blood pres- sure: findings of the Chicago Heart Association Detection the low socioeconomic status group. However, Project in Industry, 1967–1972. J Chron Dis 1995;28:527– among the upper socioeconomic status, higher 48. blood pressures may be attributed in part to 2 Cottington EM, Brock BM, Mouse JS, Hawthorn VM. Psy- chosocial factors and blood pressure in the Michigan stress caused by increased responsibilities in statewide blood pressure survey. Am J Epidemiol 1985;121: providing support to extended family members 515–29. 3 James SA. Psychosocial precursors of hypertension: a review and maintaining an observable upper socioeco- of the epidemiologic evidence. Circulation 1987;76 (suppl nomic status lifestyle. This may explain the 1):160–6. 4 Anderson NB, McNeilly M, Myers J. Autonomic reactivity finding of highest blood pressures among those and hypertension in blacks: a review and proposed model. of high John Henryism and upper socioeco- Ethn Dis 1991;1:154–70. nomic status level. In this regard, other 5 Calhoun DA. Hypertension in blacks: socioeconomic stress and sympathetic nervous system activity. Am J Med Sci measures of sociodemographic related stres- 1992;304:306–11. J Epidemiol Community Health: first published as 10.1136/jech.52.3.186 on 1 March 1998. Downloaded from 190 Markovic, Bunker, Ukoli, et al

6 Syme SL, Oakes TW, Friedman GD. Social class and racial 15 McKetney EC, Ragland DR. John Henryism, education, diVerences in blood pressure. Am J Public Health and blood pressure in young adults. The CARDIA Study. 1974;4:619–23. Am J Epidemiol 1996;143:787–91. 7 Marmot MG, Kogevians M, Elston MA. Social/economic 16 Bunker CH, Ukoli FAM, Nwankwo MU, et al. Factors asso- status and disease. Annu Rev Public Health 1987;8:111–37. 8 Hypertension Detection and Follow-up Program Coopera- ciated with hypertension in Nigerian civil servants. Prev tive Group. Race, education and prevalence of hyperten- Med 1992;21:710–22. sion. Am J Epidemiol 1977;106:351–61. 17 Weinrich SP, Weinrich MC, Keil JE, Gazes PD, Potter E. 9 James SA, Hartnett SA, Kalsbeek WD. John Henryismism The John Henryism and Framingham Type A Scales. and blood pressure diVerences among black men. J Behav Measurement properties in elderly blacks and whites. Am J Med 1983;6:259–77. Epidemiol 1988;128:165–78. 10 James SA, Strogatz DS, Wing SB, Ramsey DL. Socioeco- nomic status, John Henryism, and hypertension in blacks 18 Bunker CH, Ukoli FA, Matthews KA, Kriska AM, Huston and whites. Am J Epidemiol 1987;126:664–73. SL, Kuller LH. Weight threshold and blood pressure in a 11 James SA, Keenan NL, Strogatz DS, Browing SR, Garrett lean black population. Hypertension1995;26:616–23. JM. Socioeconomic status, John Henryism, and blood 19 Dischinger P, DuChene AG. Quality control aspects of pressure in black adults. The Pitt County Study. Am J Epi- blood pressure measurements in the Multiple Risk Factor demiol 1992;135:59–67. Intervention Trial. Control Clin Trials 1986;7 (suppl):137– 12 James SA. John Henryism and the health of African- 57S. Americans. Cult Med Psychiatry 1994;18:163–82. 20 Dressler WW. Social class, skin color, and arterial blood 13 Duijkers TJ, Drijver M, Kromhout D, James SA. “John pressure in two societies. 1991;1:60–77. Henryism” and blood pressure in a Dutch population. Psy- Ethn Dis chosom Med 1988;50:353–9. 21 Markovic N, Olomu IN, Bunker CH, Huston SL, Ukoli 14 Light KC, Brownley DA, Turner R, et al. Job status and FAM, Kuller LH. Adequacy of a single visit for high-eVort coping influence work blood pressure in women classification of hypertensive status in a Nigerian civil serv- and blacks. Hypertension 1995;25 (part 1):554–9. ant population. Int J Epidemiol 1994;23:723–9. http://jech.bmj.com/ on September 27, 2021 by guest. Protected copyright.