650 Module 8: Activities

377

Journal of Occupational and Organizational Psychology (2002). 75, 377-392

© 2 0 0 2 The British Psychological Society

www.bps.org.uk

Readiness for organizational change: A longitudinal study of workplace, psychological and behavioural correlates

Charles E. Cunningham'*, Christel A. Woodward^, Harry S. Shannon^, John Macintosh', Bonnie Lendrum', David Rosenbloom' and Judy Brown^ 'Hamilton Health Sciences and McMaster University, Hamilton, Ontario, Canada ^McMaster University, Hamilton, Ontario, Canada

To examine factors influencing readiness for healthcare organizational change, 654 randomly selected hospital staff completed questionnaires measuring the logistical and occupational risks of change, ability to cope with change and to solve job- related problems, social support, measures of Karasek's (1979) active vs. passive job construct (job demand x decision latitude) and readiness for organizational change. Workers in active jobs (Karasek, 1979) v/hich afforded higher decision latitude and control over challenging tasks reported a higher readiness for organizational change scores. Workers with an active approach to job problem-solving with higher job change self-efficacy scores reported a higher readiness for change. In hierarchical regression analyses, active jobs, an active job problem-solving style and job-change self-efficacy contributed independently to the prediction of readiness for organiz- ational change. Time I readiness for organizational change scores and an active approach to job problem-solving were the best predictors of participation in redesign activities during a year-long re-engineering programme.

Healthcare organizations are undergoing unprecedented changes (Shortell, Gillies, Anderson, Erickson, & Mitchell, 1996). (Competition, funding reductions, efforts to improve cost-efficiency, mergers and the re-engineering of work processes are placing enormous demands on healthcare organizations and their employees (Woodward et al., 1999). Research on individual differences in readiness for organizational change, workplace processes that facilitate change and factors that influence the impact of organizational change on the health and emotional well-being of employees Is important to the success of efforts to improve the health service delivery system.

Readiness for change research suggests that a demonstrable need for change, a sense of one s ability to successfully accomplish change (self-efficacy) and an opportunity to participate in the change process contribute to readiness for organizational change (Armenakis, Harris, & Mossholder, 1993). Readiness for change models have been applied widely in the organizational and behavioural sciences. Prochaska and

^Requests for reprints should be addressed to Charles E. Cunningham, Department of Psychiatry and Behavioural Neurosdences. Faculty of Health Sciences. McMaster University, 1200 Main Street, Homiltor], Ontario, L8N 3Z5. Canada (e-mail: [email protected]).

378 Charles E. Cunningham et al.

colleagues, for example, found that readiness for individual change proceeded through stages (Prochaska et al., 1994; Prochaska, Redding, & Evers, 1997) beginning at the precontemplative stage, where the need for change is not acknowledged. At the contemplative stage, individuals consider but do not initiate change. As a preparatory stage is reached, planning for change occurs (Prochaska etal., 1994, 1997). Individuals engaged in the process of behavioural change are at the action stage, whereas those attempting to sustain changes are at the maintenance stage. Movement through these stages is governed by decisional balance, the anticipated risks of change vs. the potential benefits of change.

This study applied an individual readiness for change model to a longitudinal study of organizational re-engineering in healthcare settings. We developed a brief measure of individual readiness for organizational change based on Prochaska et a/.'s (1994) questionnaires and tested several assumptions of individual and organizational readiness for change models.

Benefits vs. risks of organizational change Readiness for change begins with an individual's perception of the benefits of change (Prochaska et al., 1994), the risks of failing to change (Armenakis et al., 1993; Beer, 1980; B. A. Spector, 1989), or the demands of externally imposed changes (Pettigrew, 1987). We hypothesized, therefore, that workers' perceptions of opportunities for improvement in staff competence, service quality, quality improvement programme or organizational staff relationships would contribute to readiness for change scores.

Readiness for change research suggests that staff perceptions regarding the risks of re-engineering should also influence readiness for organizational change (Prochaska et al., 1994). Employees in healthcare organizations facing re-engineering are con- fronted with at least three types of risks. First, because organizational re-engineering poses a threat of job change or loss, we postulated that perceptions of occupational insecurity would lower readiness for organizational change scores and limit partici- pation in re-engineering activities. Second, the logistical burden of re-engineering represents a risk that may shift decisional balance and reduce readiness for organiz- ational change (Prochaska et al., 1997). In a healthcare work force composed largely of women, the domestic responsibilities assumed by many employees might increase the logistical demands of organizational re-engineering (Hall, 1989, 1992). Shift work might contribute to an already difficult logistical burden. We predicted, therefore, that child care, household tasks, shift work and a perception of conflict between domestic and occupational responsibilities would reduce readiness for organizational change and limit participation in re-engineering. Third, because organizational change represents a considerable source of stress (Eerrie, Shipley, Marmot, Stansfeld, & Smith, 1995; Woodard et al., 1999), re-engineering may pose special risks lor employees experiencing psychological distress. We hypothesized that emotional exhaustion and depression would reduce readiness for organizational change and participation in redesign activities.

Individual contributors to readiness for organizational change Self-efficacy, the perceived ability to manage change successfully, exerts a mediating effect on readiness for individual (Prochaska et al., 1997) and organizational change (Armenakis et al., 1993; Pond, Armenakis, & Green, 1984). Workers with confi- dence in their ability to cope with change should be more likely to contribute to

Readiness for organizational change 379

organizational redesign. In contrast, workers may resist cbanges that they believe exceed their coping capabilities (Armenakis et al., 1993; Bandum, 1982). We pre- dicted, therefore, tbat staff who were cotifidetit in their ability to cope with job change and who adopted ati active approach to job problem-solvitig would bave a higher readiness for cbatige scores and participate in a greater number of organizational redesign activities.

Workplace contributors to readiness for organizational change Finally, we a.ssumed that individual readiness for cbange would be influenced by broader organizational factors. Jobs wbich empower (Spreitzer, 1995) employees with the skills, attitudes and opportunities to manage change should increase work-related self-efficacy (Conger & Kanungo, 1988) and readiness for organizational change (Armenakis et al., 1993; Neuman, 1989). Karasek (1979) described active jobs as psycbologically demanding positions affording bigh decision latitude. Jobs with low demands and loŵ decision latitude were defined as passive (Karasek, 1979)- Active jobs increase learning opportunities and contribute to desirable stress, whicb increases motivation and tbe development of new bebaviour patterns (Tbeorell & Karasek, 1996). Active jobs provide opportunities for enactive mastery and incremental prep- anition for larger-scale organizational cbange (Armenakis et al., 1993). Workers in active jobs should be more confident in their ability to manage cbange (Spreitzer, 1995) and better prepared to participate in organizational redesign (Armenakis et al.., 1993; Beer & Walton, 1987; Neuman, 1989). Passive jobs, wliicb limit opportunities for decision-making and control, may compound tbe anticipated occupational risks of organizational re-engineering, lower self-efficacy and limit readiness for change. A third factor, .social support (Johnson, 1991; Karasek, Triantis, & C ĥaudhry, 1982; LaRocco, House, & French, 1980; Stansfeld, North, White, & Marmot, 1995) appears to interact witb active jobs to predict workplace adjustment. We predicted, therefore, that higher scores on both Karasek and Tbeorell's (1990) active vs. passive job dimension and social support would be associated witb readiness for organizatiotial cbange and stibsequent participation in a year-long hospital re-engineering programtne.

Method

Participants A sample ot 880 staff, =21% of the employees at a large Canadian teaching hospital, was randomly selected from tbe organization s human resources files. Participants were drawn from a wide range of job descriptions (e.g. nurses, pbysiotherapists, housekeep- ing) at two former general hospital sites wbich eacb possessed a ntimber of units (a cbildren s bospital, a cbildren's outpatient developmental and mental bealth centre, a rehabilitation hospital and a chrotiic care setting).

Procedures Baseline surveys were sent to staff selected for tbe study after the intent to re-engineer was amioimced, but several months before redesign planning began. Staff were itiformed tbat the purpose of the stirvey was to understand bow workplace cbanges affect both employees and services. Staff returned questionnaires to a university research unit witb assurance tbat data would remain confidential, and tbe hospital admitiistnition would not have access to individual scores. Following baseline sitrveys,

380 Charles E. Cunningham et al.

an extensive programme of organizational re-engineering began. Design teams worked for a year to achieve cost reductions and service improvements by introducing pro- gramme management, designing evidenced-based clinical pathways, redistributing tasks (multiskilling) and reducing staff. Staff were informed by regular newsletters and town hall discussions and encouraged to pose questions, make suggestions or offer feedback anonymously. Staff were given opportunities to participate in a wide range of redesign activities and provided with transitional workshops and supports (e.g. resume preparation, interview training, computer skills and career planning). At Time 2, one year following the completion of the first survey, the same cohort of employees was sent a second survey.

Measures

Measures of the risks ofchatige Family demographics. Participants completed questions regarding marital status, number of chiitlren, time devoted to child care, care of extended family members and family income.

job insecurity. A 6-item 5-point (strongly disagree to strongly agree} scale (alpha=.65) measured job insecurity (Greenhalgh & Rosenhlatt. 1984).

job ir)terference. A 3-item 3-point (not at all to a great deal) scale (alpha=.64) from the Whitehall studies measured the adverse effects of one's job on family life (North etal., 1993).

tAeasures of self-efficacy

jotxhange seif-efficacy. Confidence in one's ability to cope with job change, the transfer- ability of joh skills, and joh prospects were measured with a new 5-item 5-point (strongly disagree to strongly agree) seale (alpha= .71).

Active approach to job problem-soiving. This 5-item 5-point (never to almost all the time) scale, which measured an active approach to the solution of work-related prohlems (alpha=.75), was adapted from Israel, Schurman, and House (1986).

Aleasures of joh characteristics

job demands. A 6-point 5-item (strongly disagree to strongly agree) variation of the original 4-point scale (Karasek, 1985) measured psychological and physical job demands (alpha = .69).

Decision latitude. A 9-item 5-point (strongly disagree to strongly agree) variation of the original 4-point scale (Karasek, 1985) composed of skill discretion (the breadth of skills workers could use) and decision-making authority measured decision latitude (alphas.60).

Active vs. passive job. Karasek's (1979) active joh construct was computed hy multiplying decision latitude by job-demand scores.

Social support. A 10-item 5-point (strongly disagree to strongly agree) scale (alpha=.86) composed of three supervisor support and seven colleague support

Readiness for organizational change 381

questions was adapted from Karasek's Job Content Questionnaire (four new and six adapted questions).

Organizationlstaff relations. This 5-item 5-point (poor to excellent) scale (alpha=.79) was adapted from the Hospital Corporation of America's staff judgments of hospitals questionnaire (Hays, 1994). For example, staff rated 'The way this hospital treats its employees'.

Service quality Quality of patient core. A l6-item 5-point (poor to excellent or don't know) scale (alphas .93) was adapted from the Hospital Corporation of America's quality of patient care measures (Hays, 1994). For example, staff rated the 'current quality of care provided'.

Staff competence. This 9-item 5-point (poor to excellent or don't know) scale (alpha=.94) included items from the Hospital Corporation of America's measures (Hays, 1994). For example, staff rated the extent courtesy and respect are shown patients by staff.

Attention to quality improvemer^t This measure was a 5-item, 5-point (poor to excellent to don't know) scale based on the Hospital Corporation of America's questionnaire (Hays, 1994) plus additional items examining staff perceptions of the hospital's com- mitment to quality improvement (alpha=.88). For example, staff rated the emphasis placed on evidence to guide improvement of quality in eare and services'.

Psychological measures Readiness for organizational change. A 6-item 5-point (strongly disagree to strongly agree) readiness for change scale (alpha=.63) was modelled after the measures developed by Prochaska and colleagues (1994), with questions reflecting the precontemplative, contemplative, preparatory, action and maintenance stages of the model (Prochaska et al., 1994). Scoring for items at the precontemplative stage (e.g. programme does not need changing) were reversed to yield a continuous scale, with higher scores reflecting increased readiness.

Planned health and lifestyle changes. Staff checked lifestyle-change options from a list of 15 Ontario Health Survey items (Ontario Ministry of Health, 1992).

Emotional exhaustion. Staff completed the 7-item 6-point (never to every day) Emotional Exhaustion seale (alpha= .91) from the Maslaeh Burnout Inventory (Maslach & Jackson, 1981).

Depression. Symptoms of depression ^vere measured with a 10-item 4-point (rarely or none of tbe time to mostly all of tbe time) scale (alpha=.78) version of the Centre for the Epidemiological Study of Depression scale (CESD) (Radloff, 1977; Reis & Herz, 1986).

Contributions to re-engineering Participation in re-engineering. At Time 2, one year after the start of redesign, staff recorded which of seven possible re-engineering activities (e.g. submitting design ideas, volunteering for design teams, joining design teams, participating in redesign

2.9

3.5

3.5

3.5

1.0

1.0

I.I

1.2

382 Charles £ Cunningham et al.

Table I . Mean and standard deviations for readiness for organizational change questions"

Stage of change Content of question Mean SD

Precontemplative stage The programme or area in which I work functions well and does not have any

aspects which need changing There's nothing that I really need to change about the way I do my job to be

more efficient Contemplative stage I've been thinking that I migbt want to help change something about the

programme or area in which I work Preparatory stage I plan to be involved in changing the programme or area in which I work A c t i o n ^ tage I am working hard to help improve aspects of the programme or area in which I

work 3.7 I.I Maint^ance stage W e are trying to make sure we keep changes/improvements my programme/area

has made 3.5 .9

"Scale range I (strongly disagree) t o 5 {strongly agree).

work groups, etc.) they participated in during the year following Time 1 assessments. At Time 2, staff rated their contribution to the re-engineering process on a 5-point scale (not at all to very much).

Results This analysis begins by examining the workplace, psychological and domestic correlates of baseline readiness for org:mizational change. Hierarchical regression equations (Tabachnik & Fidel, 1996) were computed to model the contribution ol the anticipated risks of change, job change sef-efficacy, an active approach to job-related problem-solving and Karasek's (1979) model of an active job to readiness for organiz- ational change. A final regre.ssion equation examined the predictors of participation in a year-long programme of organizational re-engineering. Responses to the six items on the readiness for organizational change measure are presented in Table 1.

Response rate Of the 65^ (74'X)) staff returning surveys, most were women (87%) mth community college (44%) or university degrees (28%). Most participants were nonsupervisory (87%), nonunion (87%) staff, employed on hourly contracts (77%) with shift work requirements (44%). A majority of participants were married or living with partners (76%) and had children in the home (71%). Although participants were employed an average of 35 h per week, a considerable amount of additional time was devoted to child care (29 h) and household activities (14.7 h). Of the 834 employees eligible for participation at Time 2, 528 (63%) completed surveys.

Risks and benefits of organizational change Correlational analysis showed that domestic factors which might influence the logisti- cal burden of change, or limit participation in the redesign process, were not linked to

Readiness for organizational change 383

readiness for organizational change scores. With the exception of children under age 6 in the home, whicb was related to lower participation (r=-.l6**) but not readiness (r=-.O7), gender (r=.O2, r=.O5), time devoted to child care (r=-.O3, r=-.O7), responsibility as the main family earner (r= - .08, r= - .04), household chores (r= - .08, r=-.O5), and marital status (r=.03. r=.O3) were not related to readiness for change or participation in redesign activities, respectively. Table 2 shows that staff with a higher readiness for organizational change scores reported slightly more job interference with family life and higher emotional exhaustion scores. Shift work, which might limit participation in redesign activities, was associated with a slightly lower readiness for organizational change scores (r=-.15**) and less participation in redesign activities (r=-.25**). Job insecurity (Table 2) was not linked to readiness for organizational change or participation in redesign activities.

The correlation analysis showed that readiness for organizational change scores were not linked to potential benefits of change. Staff perceptions of the quality of patient care (r=-.O4, r=-.O4), statT competence ir=-.O6, r=-,09), quality improve- ment programmes (r=.OO, r=-.O7), or staff-organizational relations (r=.O8, r=.O4) were not linked to readiness for organizational change or participation in redesign activities, respectively.

Individual correlates of readiness for organizational change As predicted, Table 2 shows that workers with an active approach to job related problem-solving and higher job-change self-eftlcacy at Time 1 reported a higher readi- ness for change scores, participated in a greater number of redesign activities during the following year, and reported making a greater contribution to organizational change.

Workplace contributors to readiness for organizational change Table 2 shows that staff in active jobs (e.g. high decision latitude x high job demands) reported a higher readiness for organizational change scores, participated in a greater number of redesign activities, and felt they made a greater contribution to organiz- ational change than those in passive jobs. Although social support was only weakly related to readiness for organizational change and was not related to participation in redesign activities, it was associated with lower emotional exhaustion scores.

As a measure of discriminant validity, we detemiined whether readiness for organiz- ational change was specific to the workplace or reflective of a more generalized readiness for personal change. Readiness for change was not linked to plans to engage in health-related personal lifestyle changes (r=.()6).

As shown in Table 2, staff with a higher readiness for organizational change scores at Time 1 participated in a greater number of redesign activities during a year-long re-engineering programme and reported making a greater contribution to organiz- ational change at Time 2.

Predicting readiness for organizational change A hierarchical regression equation (Tabachnik & Fidell, 1996) tested the predictions of individual (Prochaska et al., 1994, 1997) and organizational readiness for change models (Armenakis et al., 1993). Prochaska s model suggests that readiness reflects a balance between the risks and benefits of change. After controlling for hospital site on step 1, step 2 entered occupational and logistical risks of change: shift work, job

384 Charles E. Cunningham et ai

o U

re

- r — —

I I

t I o o —

* 4. ro O (N — _ O

o o 00 LTl 00

• o

r-..

1= u H ---

a. U

5 "

O Q < i/i

o

Readiness for organizational charige 385

Table 3. Sequential regression equation predicting readiness for organizational change scores (N=62S)

Measure

Step 1: Control variable Site

Step 2: Occupational risks Shift work Job interferes with family Job insecurities

Step 3: Psychological risks Emotional exhaustion Depression

Step 4: Self-efficacy Job change self-efficacy Active job problem-solving approach

Step 5: Karsek's active job Active vs. passive job

M u l t r

.12

11

.31

.48

.54

Mult r̂

.02

.07

.10

.23

.29

2 change

.02

.06

.02

.13

.07

Beta''

.06

- . 1 1 .11 .12

.03 - . 1 0

.14

.26

.29

t

1.83

- 3 . 1 5 * * 2.36* 2.81**

.60 - 2 . 5 1 * *

3.20*** 7.03***

7 55***

"Standardized beta from final step of the regression equation.

interference with family and job insecurity'. On step 3, ^ve entered psychological variables that might increase the risk of change: emotional exhaustion and depression. Because readiness for organizational change is mediated by self-efficacy, the percep- tion that one can manage change (Aremenakis et al., 1993; Prochaska et al., 1994), measures of job change self-efficacy and an active approach to job problem-solving were entered on step 4. Finally, because an active job should prepare workers for organizational change (Armenakis et al., 1993; Spreitzer, 1995; Theorell & Karasek, 1996), we entered Karasek's active vs. passive job construct (decision latitudexjob demands) on step 5.

Table 3 shows the multiple r, multiple R^, R^ change, standardized regression coefficients, f-values and probability levels following entry of each variable. Overall, this model accounted for 29% of the variance in readiness for organizational change scores. Standardized beta scores for the final step showed that an active job and an active approach to job problem-solving were tbe best predictors of readiness for organizational change scores. Entering the components of Karasek's active vs. passive job construct, decision latitude and job demands before active vs. passive job in a second hierarchical regression equation did not contribute significantly to the prediction of readiness for organizational change.

Predicting participation in organizational change We postulated that, in combination with anticipated risks, self-efficacy and active jobs, baseline readiness for change scores would predict participation in a year-long process of organizational redesign. The hierarchical regression equation summarized in Table 4 shows that this model accounted for 27% of the variance in re-engineering partici- pation. Standardized beta scores for the final step suggest that Time I readiness for change scores and an active approach to job problem-solving were the best predictors of participation in redesign activities. Shift work and higher depression scores were

386 Charles E. Cunningham et al.

Table 4. Hierarchical regression equation; Time I variables predicting Time 2 participation in

redesign activities (N=450)

Measure

Step 1: Control variable Site

Step 2: Occupational risks Shift v^ork Job interferes with family Job insecurity

Step 3: Psychological risks Emotional exhaustion Depression

Step 4: Self-efficacy Job change self-efficacy Active job problem-solving approach

Step 5: Karsek's active job Active vs. passive job

Step 6: Readiness for change Time 1 readiness for organizational change

M u l t r

.10

.31

.35

.47

.48

.52

Mult r̂

.01

.10

.12

.22

.23

.27

J change

.01

.09

.03

.10

.01

.04

Beta"

.06

- . 2 0 .11 .05

.08 - . 1 1

- . 0 1

.23

.03

.24

t

1.32

_ 4^9*^M<

2.02̂ *̂

.90

1.24 - 2.29*

- . 1 3 4 . 9 5 * * *

.61

4.88***

"Standardized beta from final step of the regression equation.

associated with lower participation in redesign activities. Entering the compotients of Karaseks active vs. passive job construct, decision latitude and job demands before active vs. passive job in a second hierarchical regression equation did not contribute significantly to the prediction of participation in redesign activities.

lo determine whether participation was a function of the opportunities afforded by supervisory status, we computed the same sequential regression equation in a subsaniple of 274 nonsupervisor>' health professit)nals (Table 5). This model, again, accounted for 27% of the variance in participation in redesign activities. Significant beta scores for the final step suggest that a higher readiness for change scores, an active approach to job problem-solving and site were the best predictors of partici- pation in redesign activities. Shift work was, again, associated with less involvement in redesign activities.

Discussion

Organizational contributors to readiness for organizational change Readiness for change was best predicted by combining organizational (Armenakis etal., 1993) and individual models (Prochaska etal., 1994, 1997). Work variables were the best predictors of readiness for organizational change. Employees in active pos- itions witb more control over challenging jobs reported a higher readiness for organiz- ational change sct)res dmi were mt)re likely to participate in organizational redesign. This is consistent with research suggesting that active jobs foster personal empower- ment, improve performance, increase initiative and contribute to organizational innovation (Conger & Kasungo, 1988; Spreitzer, 1995; fheorell & Karasek, 1996).

Readiness for organizational change 387

Table 5. Sequential regression equation of baseline measures predicting participation in organiz- ational re-engineering during the next year

Measure

Step I: Occupatfonal risks Job insecurity

Step 2: Logistical risks Shift v/ork Job interferes with family

Step 3: Psychological risks W o r k stress Emotional exhaustion Depression

Step 4: Self-efficacy Job change self-efficacy Job problem-solving

Step 5: Active job Step 6: Readiness for change

M u l t r

.030

.292

.353

.469

.491

.522

Multr*

.001

.085

.124

.220

.241

.272

Beta"

.060

- . 1 8 4 .096

- . 0 3 8 .052

- . 1 1 2

- . 0 1 6 .205 .307 .215

t

I.I

- 4 . 2 1 1.80

- . 7 5 .83

- 2 . 2 9

- . 2 9 4.33 7.8 4.35

P

.272

.012

.073

.453

.408

.022

<.OOI <.OOI <.OOI <.OOI

"Standardized beta from final step of the regression equacion.

The dynamic demand-control hypothesis suggests that active jobs contribute to a sense of mastery (Theorell & Karasek, 1996). In the present study, positive correlations among active jobs, an active approach to job problem-solving and bigber job cbange self-efficacy are consistent with these predictions. Theorell and Karasek (1996) also suggested that the sense of mastery created by active jobs inhibits the perception of stress or engenders positive stress (Theorell & Karasek, 1996). Active jobs, however, were associated with greater interference with family and higher emotional exhaustion scores. This observation questions the assumption that active jobs contribute to a state of positive stress.

Previous studies have linked workplace social support to employee adjustment Oohnson, 1991; Karasek et al., 1982; LaRocco et al., 1980; Stansfeld et al, 1995). In this study, job-related interpersonal relationships made a very limited contribution to the prediction of readiness for organizational change scores. A socially supportive workplace, however, was correlated with lower emotional exhaustion scores. These findings suggest that supportive colleagues may play a more important role in employee efforts to cope with the stress of organizational change (Woodward et al., 1999).

Decisional balance: Risks and benefits of change Readiness for change models suggest that evidence of a need for change, a discrepancy between present conditions and a targeted organizational objective, are important to the creation of readiness for organizational change (Armenakis et al., 1993; Beer, 1980; B. A. Spector, 1989). As in many healthcare redesign initiatives (Ho, C:han, & Kidwell, 1999), the need to change was imposed by funding reductions. In this study, staff judgments of the quality of the care and services provided by the organization were not linked to readiness for change scores. When imposed change represents occupational, logistical and psychological risks to employees, and is not linked to a

388 Charles £, Cunningham et al.

perceived need for quality improvements, the success of these changes may be compromised (Armenakis et al., 1993; Ho et al., 1999; B. A, Spector, 1989),

Decisional balance models suggest that individuals prepare for action when the perceived benefits of change outweigli the anticipated risks of change (Prochaska et al., 1994). Our results provide limited support for this model. Thus, shift work, which poses logistical risks to employees considering participation in organizational redesign, was linked to a lower readiness for organization;il change scores and less participation in re-engineering activities.

Most of the employees studied bere were women. The child care and household responsibilities which employed women assume may prolong job-related physiological arousal, intensify work overload and role conflict, and increase vulnerability to workplace stress (Burke, 1993; Eckcnrode & Gore, 1990; Hall, 1989, 1992; Keita & Jones, 1990). In this study, the perception th;it work interferes with family life was strongly linked to emotional exliaustion and depression scores. Staff reporting a higher readiness lor change scores and participating iii a greater number or redesign activities reported a slightly greater interference with their family and somewhat higher emotional exhaustion scores. Nonetheless, children in the home under age 6 was the only domestic responsibility measure linked to lower participation in redesign activities.

Organizational redesign is a stressful experience for many staif (Woodw:ird et al., 1999). Nonetheless, psychological factors, which might increase the personal risks of rapid organizational change, did not reduce readiness for change. As noted above, workers with a higher readiness for change scores reported slightly higher emotional exhaustion scores. This is consistent with studies finding that organizational stressors may prompt innovation and a coping response (Bunce & West, 1994). Finally, contrary to the predictions of decisional balance models, a key work-related measure of risk, job insecurity, was not related to readiness for organizational change or participation in redesign activities.

Individual correlates of readiness for organizational change Self efficacy, the perceived ability to cope with change, is thought to be an important contributor to readiness (Armenakis et al., 1993; Pond et al., 1984; Prochaska et al., 1994), In this study, workers with an active approach to work problems, who were more confident in their ability to cope with job change, reported a higher readiness for organizational change scores at Time 1, participated in a greater number of redesign activities during the following year, and felt that they made a greater contribution to organizational change at Time 2, These data are consistent with studies on personal hardiness suggesting that a similar constellation of psychological characteristics is associated with favourable responses to stressful events (Kobasa, Maddi, Fucetti, & Zola, 1985; Kobasa &Pucetti, 1983; Westman, 1990).

Measuring readiness far organizational change In tliis study, statf with higher readiness scores participated in more re-engineering activities and felt that they made a greater contribution to the organization s redesign efforts. The data support the predictive validity of this measure. The independent contribution of readiness for change scores to the predication of participation provides iurther support for this measure's predictive validity.

Readiness for organizational change 389

The finding that readiness for organizational change scores were not related to the planning of personal health behaviour changes supports the discriminant validity of this measure. Readiness for organizational change scores did not reflect a general propensity to personal change.

Limitations

Although a single corporate sampling frame may have limited the generalizability of these findings, participants were selected from different occupational groupings, inpatient and outpatient settings, and geographical locations. Participants worked in programmes serving children and adults with a wide range of acute and chronic health problems at sites with different funding bases, histories and ctiltures.

In addition, estimates of decision latitude, social support and individual readiness for organizational change were based on employee reports. Although subject to informant biases, perceptions of workplace eharaeteristics have been linked to more objective analyses (P. Spector, Dwyer, & Jex, 1988). Moreover, physiological responses are mediated by the interaction of perceptions of workplace characteristics (e.g. job control) and more objective measures (Fox, Dwyer, & Ganster, 1993; Stansfeld et al., 1995).

Implications

This study has several implications for healthcare organizations facing change. First, active jobs that afford control over challenging tasks, conditions which optimize health and emotional well-being (Blegen, 1993; Greenberger, Strasser, Cummings, & Dunham, 1989; Karasek & Theorell, 1990; Landsbergis, 1988; Tetrick & LaRoeco, 1987; Tbeorell & Karasek, 1996), prepare workers to initiate or contribute to organizational change.

Second, encouraging an active approach to job problem-solving, building the strategies needed to manage change suceessfully and enhancing job ehange self- effieacy (Armenakis et «/., 1993) should contribute independently to the preparatory benefits of an active job.

Third, workers in demanding jobs reported a higher emotional exhaustion. Longi- tudinal studies suggest that organizational re-engineering increases emotional exliaus- tion (Woodward et al., 1999), wbich may adversely effect work performance (Wright & Bonett, 1997) and lower patient satisfaction (Leiter, Harvie, & Frizzell, 1998). If the strain of rapid organizational change is not modulated, the contribtitions of employees, and the ultimate impact of redesign on patient care, may be compromised. Building supportive relationships with co-workers and supervisors and limiting confiiet between work and home life might modulate the stress of organizational change.

Stage models suggest that workers at the precontemplative or contemplative stages would respond to different interventions to those at the preparatory or action stages (Prochaska et al., 1994, 1997). Studies linking organizational functioning to work- group readiness for ehange suggest that employees in the same work group would be at similar stages (Fox, Ellison, & Keith, 1988). The types of brief measures used here may help identify workgroup readiness levels and design preparatory interventions addressing the needs of employees at different stages of the change process.

Stage models suggest a dynamic process of individual and organizational change (Prochaska et al., 1994, 1997) in which a shift in the perceived risks and benefits of change may prompt return to an earlier phase. This study suggests that active involve- ment in organizational ehange, reducing barriers to participation (e.g. shift work).

390 Charles E. Cunningham et al.

building problem-solving strategies and enhancing workers" perceptions of their ability to cope witb change (change sclf-efficacy) should botb enhance commitment to redesign and reduce the stress of organizational change.

Acknowledgements

This research was supported by Haniiltoti (Icalth Sciences and the Social Sciences and Humanities Research Foundation during the collection of this data. Preparation of this manuscript was supported by a Senior Research Fellowship to the first author from the Ontario Mental Health Foundation and the John C. Laidlaw Chair in Patient Centred Health Care. Deborah Fitzpatrick provided helpful comments during the preparation of the manuscript.

References

Armenakis. A, A,, Harris, S, G.. & Mossholder, K. W, (1993). Creating readiness for organizational change. Human Relations, 46, 681-703.

Ilandura, A, A, (i982), Selt-efficacy mechanism in human agency, American Ps}>chologist, 37, 122-147,

Beer, M. (t980). Organization change and dei'elopment: A systems vieu: Santa Monica, CA: Goodyear,

lleer, M,, & Walton, A. E, (1987). Organization change and development. Annual Review of Psychology, 38, 339-367.

Blegen. M. A. (1993), Nurses' job satisfaction: A meta-analysis of related variables. Nursing Research. 42, 36-41,

Bunce, I),, & West, M, (1994), c:hanging work environments: Intiovative coping responses to occupational stress. Work and Stress, 8, 319-331.

Burke, R. J. (1993). Organizational-level interventions to reduee occupational stre.ssors. Work and Stress, 7, 77-87.

Conger, J. A., & Kaniingo, R. N. (1988), The empowerment process: integrating theory and pntctice. Academy of Management Review, 13, 471-482,

Eckenrode, .)., & Gore, S, (1990). Stress and coping at the boundary' of work, aiid family. In J. Eekenrode & S. Gore (Eds,), Stress between ivork and family (pp. l-l6). New York: Plenum Press.

Ferrie.J. E., Shipley, M. J., Marmot, M. G., Stansfeld, S., & Smith, G. D. (1995), Health effects of anticipation of job change and non-employment: Longitudinal data from the Wbitehall II study, British MedicalJtnirnal. 311, 1264-1269.

Fox, M, I.,, Dwyer, I). J,. & Clanster, D. C, (1993), Effects of stressful job demands and control on physiological and attitudinal outcomes in a hospital setting. Academy of Management Review, 30, 289-318.

Fox, D. G., Ellison, R, L, & Keith, K, L. (1988). Human resource management: An index and its relationship to readiness for change. Public Personnel Management, 17, 297-302,

Cireenberger, D, B., Strasser, S., Cummings, L. S., & Dunham, R. (1989), The impact ol personal control on pertbrmance and satisfaction. Organizational Behavior and Human Decision Processes, 43, 29-51.

Greenhalgh, L., & Rosenblau, Z. (19H4). Job insecuritj: Toward conceptual clarity. Academy of Management Review. 9. 438-448,

Hall. E. M, (1989), Gender, work control, and stress: A theoretical discussion and an empirical test. International Journal of Health Services. 19, 72S-74S,

Hall, E. M. (1992), Double exposure: The combined impact of tbe bome and work environments on mental strain and physical illness. International Journal of Health Services, 22, 239-260.

Hays. R, D, (1994), Short from measures of physician and employee judgements about hospital quality./omr Commission Journal of Quality Improvement, 20, i^(^-ll.

Readiness for organizational change 391

Ho, S, J., Chan, L,, & Kidwell, R, E, Jr. (1999). The improvement of business process reengineer- ing in American and Canadian hospitals. Health Care Management Review, 24, 19-31.

Israel, B. A.. Schurman, S, J,, & House, J. A, (1986). Stress and wellness project survey. Ann Arbor, MI: University of Michigan, School of Public Health,

Johnson. J, V. (1991), Collective control: strategies for survival in the workplace, InJ, V. Johnson & G, Johansson (Eds,), The psychosociat work environment: Work organization, democra- tization, and health: Essays in memory of Berbil Gardell (pp, 121-132). Amityville, NY: Baywood.

Karasek, R. A, (1979), Job demands, job decision latitude, and mental strain: Implications for job redesign. Administration Science Quarterly, 224, 285-307,

Karasek, R. A. (1985)- Job Content Instrument Questionnaire (Mimeograph). Lowell, MA: University of Massachusetts.

Karasek, R, A,, & Theorell, T. (1990), Healthy work—Stress, productivity, and the reconstruction of working life. New York: Basic Books,

Karasek, R, A,, Triantis, K. P,, & Chaudhry, S. S, (1982), Coworker and supervisor support as moderators of associations between task characteristics and mental strain. Journal of Occupational Behavior, J, 181-200,

Keita, G, P., & Jones, J, M, (1990). Reducing adverse reaction to stress in the workplace, American Psychologist, 45, 1137-1141.

Kobasa, S, C, Maddi, S, R.. Pucetti, M., & Zola. M, (1985). Effectiveness of hardiness, exercise, and social support as resources against illness. Journal of Psychosomatic Research, 29, 525-533,

Kobasa. S. C, & Pucetti, M, C, (1983). Personality and social resources in stress resistance. Journal ojPersonality and Social Psychology, 45, 839-850.

LaRocco, J, M., House, J. S., & French, J, R, P, (1980). Social support, occupational stress, and hcdXth. Journal of Health and Social Behavior, 21, 202-216,

Leiter, M, P., Harvie, P,, & Erizzell, C, (1998). The correspondence of patient satisfaction and nurse humout. Social Science in Medicine, 47, 1611-1617,

Landsbergis, P, A. (1988). Occupational stress among health care workers: A test of the job demands-control model,/oMrn«/ of Organizational Behavior, 9, 217-239.

Maslach, C. & Jackson. S, E, (1981). Measurement of experienced burnout. Journal of Occupational Behavior, 2, 99-113,

Neuman, J. E, (1989). Why people don't participate in organizational change. Research in Organizational Change and Development, 3, 181-212.

North. E,, Syme, S, L, Eeeney, A., Head. J.. Shipley, M, J., & Marmot, M, G, (1993). Explaining socioeconomic differences in sickness absence: The Wliitehall II study. British Medical Journal, .iO6, 361-366,

Ontario Ministry' of Health (1992), Ontario Health Survey 1990: Users guide Vol. I documentation. Toronto: Premier s Council on Health Weil-Being and Social Justice.

Pettigrev ,̂ A. (1987). Context and action in transforming the Urm. Journal of Management Studies, 24, 649-670,

Pond, S. G,, Armenakis, A, A., & Green, S, B. (1984), The importance of employee expectation in organizational diagnosis. The Journal of Applied Behavioral Science, 20, 167-180,

Prochaska, J. O., Redding. C, A,, & Evers, K. (1997), The transtheoretical model of change. In K. Glanz, E. M, Lewis & B. K. Rimer (Eds), Health behewior and health education: Theory, research, and practice (pp. 60-84), San Erancisco: Jossey-Bass,

Prochaska, J. O,, Velicer, W. E,. Rossi, J. S., Goldstein, M. G,. Marcus, B, H,, RaJiowski. W,, Eiore, C, Harlow, L. L., Redding. C. A,, Rosenbloom, D.. & Rossi, S. R. (1994). Stages of change and decisional balance for 12 problem behaviors. Health Psychology, 13. 39-46,

Radloff, L. S, (1977), The CES-D scale: A self-report depression scale for research in the general populdtkm. Applied Psychological Measurement, /, 385-401,

Reis, J., & Herz. E. J. (1986). Factorial and discriminant validity of the Center for Epidemiologieal Studies Depression (CES-D) scale. Journal of Clinical Psychology, 42, 28-33.

392 Charles £. Cunningham et al.

ShorteU, S. M., Gillies, R. R., Anderson, D. A., Erickson, K. M., & MitcheU,.|. B. (1996). Remaking health care in America. San Francisco: Jossey-Bass.

Spector, B. A. (1989). From bogged down to fired up: Inspiring organizational change. Sloan Management Review. Summer. 29-34.

Spector, P., Dwyer, D. J., & Jex, S. M. (1988). Relation of job strcssors to affective, health and performance outcomes: A comparison of multiple data sources. Journal of Applied Psychology, 7J, H - 1 9 .

Spreitzer, G. M. (1995). Psychological empowerment in the workplace: Dimensions, measurement, and validation. Academy of Management Journal, 38, 1442-1465.

Stansfeld. S. A., North, F. M., Whitye, I., & Marmot, M. G. (1995). Work characteristics and psychiatric disorder in civil servants in lonilon. Journal of Upidemiology and Community Health, 49, 48-53.

Tabachnik, B. d.. & Fidell, L. S. (1996). Using multivariate statistics Cird ed.). New York: HarperCollins.

Tetrick, L. E., & URocco, J. M. (1987). Understanding, prediction, and control as moderators of the relationships between perceived stress, satisfaction, and psychological well-being. Jourtial of Applied Psychology, 72, 538-543.

'flieorell, T., & Karasek, R. A. (1996). Current issues relating to psychosocial job strain and cardiovascular disease research. Journal of Occupationcd Health Psychology, 1, 9-26.

Westman, M. (1990). The relationship between stress and performance: Tlie moderating effects of hardiness. Human Pcrfonnunce. , i 141-155.

Woodward, C. A., Shannon, H. S., Cunningliam, C. E., McIntosh,J. E., Lendrum, B., Rosenbloom, D., & Brown, J. (1999). The impact of re-engineering and other cost reduction strategies on the st:tff oi a large teaching hospital: A longitudinal study. Medical Care, A7, 547-555.

Wright, T. A., & Bonett, D. G. (1997). The contribution of burnout to work performance. Journal of Organizational Behavior, 18, 491-500.

Received 13 March 2000; revised version received 19 April 2002