gms | German Medical Science

GMS Psycho-Social-Medicine

Corporate journal of German scientific societies in psychosocial medicine

ISSN 1860-5214

Worrying about one's job, family, financial situation and health - results of a population-representative study

Sorgen um den Arbeitsplatz, die Familie, die finanzielle und die gesundheitliche Situation - Ergebnisse einer bevölkerungsrepräsentativen Befragung

Research Article Special issue: Diagnostic instruments

  • author Norbert Grulke - Universitätsklinik für Psychosomatische Medizin und Psychotherapie, Ulm, Germany
  • corresponding author Harald Bailer - Universitätsklinik für Psychosomatische Medizin und Psychotherapie, Ulm, Germany
  • Gerd Blaser - Universitätsklinikum Leipzig, Klinik für Psychotherapie und Psychosomatische Medizin, Leipzig, Germany
  • Michael Geyer - Universitätsklinikum Leipzig, Klinik für Psychotherapie und Psychosomatische Medizin, Leipzig, Germany
  • author Gabriele Schmutzer - Abteilung für Medizinische Psychologie, Universitätsklinikum Leipzig, Leipzig, Germany
  • author Elmar Brähler - Abteilung für Medizinische Psychologie, Universitätsklinikum Leipzig, Leipzig, Germany
  • author Cornelia Albani - Universitätsklinikum Leipzig, Klinik für Psychotherapie und Psychosomatische Medizin, Leipzig, Germany

GMS Psychosoc Med 2006;3:Doc11

The electronic version of this article is the complete one and can be found online at:

Published: December 11, 2006

© 2006 Grulke et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License ( You are free: to Share – to copy, distribute and transmit the work, provided the original author and source are credited.


Objective: Worrying about one’s job, family, financial situation and health is distressing. How intense are these worries in the general population?

Methods: An inquiry representative of the German population (N=2473, age>14 years) was performed. The total score of the “Questionnaire for Assessing Subjective Physical Well-Being“ (FEW) and the screening scale of the “Trier Inventory for the Assessment of Chronic Stress“ (TICS) were correlated with the four above mentioned items addressing worries (range 1-4).

Results: Worries about one’s financial situation were scored highest (mean=2.04, SD=0.92), followed by worries about one’s health (mean=1.94; SD=0.87), one’s family (mean=1.85; SD=0.86) and worries about one’s job (mean=1.69; SD=0.92). Health worries increased with age. Those without a job and single parents showed the highest grades of solicitude. Increasing income decreased worrying. The correlations with chronic stress were moderate (.24<r<.33). The most significant correlation could be found between worries about one’s health and the total score of the FEW (r=-.52).

Conclusions: The general population’s concerns are relatively low. Using normative data, people can be screened for psychosocial distress and possibly referred to a relevant form of psychosocial support.

Keywords: worrying, psychological health, representative inquiry, normative data


Fragestellung: Sorgen um den Arbeitsplatz, die finanzielle Situation, die Familie und die gesundheitliche Situation sind belastend. Wie hoch sind diese Sorgen in der Bevölkerung ausgeprägt?

Methoden: Bevölkerungsrepräsentative Befragung bei n=2473 deutschen Personen ab 14 Jahren. Der Gesamtwert des “Fragebogens zum körperlichen Wohlbefinden“ (FEW) und die Screeningskala des „Trierer Inventars zum chronischen Stress“ (TICS) werden für Korrelationsanalysen mit den vier Items zur Besorgtheit (Skala 1-4) herangezogen.

Ergebnisse: Finanzielle Sorgen werden im Mittel am höchsten bewertet (2,04, SD=0,92), gefolgt von Sorgen um die Gesundheit (1,94; SD=0,87), die Familie (1,85; SD=0,86) und Sorgen um den Arbeitsplatz (1,69; SD=0,92). Mit zunehmendem Lebensalter steigen die Sorgen um die gesundheitliche Situation. Die Besorgtheit ist bei Arbeitslosen alles in allem als am höchsten anzusehen, ebenfalls fallen Alleinerziehende als hoch belastet auf. Mit zunehmendem Haushaltseinkommen verringert sich der Grad der Besorgtheit. Die Korrelationen mit chronischem Stress fallen moderat aus (,24<r<,33). Der deutlichste Zusammenhang besteht zwischen Sorgen um die Gesundheit und FEW-Gesamtwert (r=-0,52).

Schlussfolgerungen: Die durchschnittliche Besorgnis der Menschen hält sich in Grenzen. Durch den jetzt möglichen Vergleich mit Normwerten können diese Fragen zum Screening psychosozialer Belastungen von Patienten verwendet werden und bieten differentielle Ansatzpunkte für deren psychosoziale Unterstützung.

Schlüsselwörter: Besorgtheit, psychisches Befinden, Repräsentativerhebung, Normdaten


Stress and worries are relevant to the development and the course of psychological disorders and probably of physical disorders as well. In clinical practice one often evaluates the extent of stress and worries using simple questions. However, no reference data from the general population exist; therefore answers to these simple questions are difficult to interpret. The present study involved a sample representative of the general population in which normative data were collected on four questions addressing worries about one’s job, family, financial and health situation. The four questions that proved to be useful in our clinical practise were similar to those asked in established instruments such as the Worry Domains Questionnaire [1] (e.g. “I worry that my money will run out.”). We do not report on a theory-driven application of the concept of worry; for example, we do not distinguish between the cognitive component of anxiety and the emotional component thereof [2].

Worries are characterized by the predominance of internal verbal cognitive activities. Worry seems like a soliloquy that focuses on the negative. Mental imaging is rare. It is human nature to worry. Worrying can serve as cognitive distraction from real problems as well as avoidance of distressing themes [3]. If worries exceed a certain level, they must be considered pathological. Meanwhile worrying is a leading symptom in generalized anxiety disorder [4]. Our study focused on “normal worrying” with respect to the mentioned four items. To our knowledge, there are no other studies focusing on “normal worrying” in the general population. The results of our study should allow for a better interpretation of answers given by patients. Further, we aimed at investigating correlations between our questions and standardized instruments measuring physical well-being and chronic stress.



- Worries

Participants were asked to assess their distress in four areas using a four-point scale:

“At present, how much do you worry about your job, your financial situation, your family, your health?” (1 = not at all; 2 = a little bit, 3 = quite a lot, 4 = very much). (Original items (German): Wie sehr machen Sie sich zurzeit Sorgen wegen: Ihres Arbeitsplatzes, Ihrer finanziellen Situation, Ihrer Familie, Ihrer gesundheitlichen Situation? (1 = „überhaupt keine“, 2 = „eher wenig“, 3 = „eher viel“, 4 = „sehr große“).

- Trier Inventory for the Assessment of Chronic Stress (“Trierer Inventar zum chronischem Stress” TICS, [5])

Different kinds of chronic distress are covered by 57 items. The authors’ concept of stress is an interactive one, stating that stress evolves and results from the interaction of a person with the requirements of the environment. Items are rated on a 5-point scale (0 = never, 4 = always). Because the reported factorial structure could not be replicated in our sample (results not reported here) we only used the screening scale of 12 items as proposed by the test authors.

- Questionnaire for Assessing Subjective Physical Well-Being (“Fragebogen zum körperlichen Wohlbefinden” (FEW [6])

This questionnaire measures the habitual subjective physical well-being of adults. Physical well-being is not defined as the absence of disease, pain or impairment but is measured using positively phrased items. The 16 items are rated on a 6-point scale (0 = definitely does not apply, 5 = definitely does apply). For our purposes, the total score (mean of all items) of the FEW is used, cf. [7].


Data sampling was performed in the context of a multipurpose survey by USUMA, an opinion research institute in Berlin, by order of Leipzig University, Germany, during autumn 2004.

Households were selected at random using the random-route procedure, the person to be interviewed was selected at random. Participants were visited and interviewed at home by trained interviewers. During the interviews participants were given the questionnaires to fill out personally. Participation was voluntary and participants were given a signed declaration of data privacy.

To ensure a representative sample ADM-samples [8] were drawn and compared to official statistics. The population in our study were Germans living in private households aged 14 years and older. The response rate was 62.3%. N=2591 persons aged 14 to 99 years participated (for socio-demographic characteristics see Table 1 [Tab. 1]).

Statistical procedures

Data were analysed using the software package SPSS for Windows (11.0.1, SPSS Inc., Chicago, IL). Differences in means were evaluated as effect sizes (difference of means divided by the pooled standard deviation). Following convention [9], effect sizes are regarded as weak (>.20), moderate (>.50) and strong (>.80). P-values <.05 were considered significant.



In general the items rated by the general population showed quite a low level of worry (see Table 2 [Tab. 2]). Worries about one’s financial situation were rated highest (mean=2.04, SD=.92), followed by worries about health (mean=1.94; SD=.87) and the family (mean=1.85; SD=.86). Worries about one’s job were rated lowest (mean=1.69; SD=.92).

For each of the four items, about 3 in 4 respondents reported worrying “not at all” or “a little” (71.9%-81.6%), and less than 10% worried “very much”.

Of all respondents, 16% reported worrying “very much” for at least one of the items and 0.5% gave this answer for all four items. 27.8% of the participants worried “not at all” for three or four items, and 50.7% worried “not at all” or “a little” for all four items.

These global results do not take relevant socio-demographic factors into account. For example pensioners and non-working people could not worry about their job. Further, it could be assumed that older persons worry more about health than younger ones. The next step analysed the effects of these factors.

Worries, age, employment

Age correlated with all four items (Pearson correlations): worrying about one’s job r=-.34, financial situation r=-.22, family r=-.05, and health r=.31 (all p<.05). In a further step the pensioners were eliminated because they did not need to be concerned about their job. By ignoring the pensioners, the mean age of the sample dropped form 48.1 to 39.9 years (SD 12.7; range 14-65 J.). The correlation coefficients diminished considerably down to: -.01, -.05, .01 und .20 in the above mentioned sequence (n=1695-1719). Due to the large sample size the coefficients of the magnitude .05 (financial situation) and .20 (health) were significant (p<.01 and <.001 respectively). The other two correlations (job and family) did not reach significance (p>.55).

As expected, the pensioners represented the oldest, and school children and trainees the youngest subgroup. The average age differed between the groups categorised with respect to employment (see Table 3 [Tab. 3]). Further, these groups differed in the extent of their worrying.

Quite conspicuous is the unemployed group. Their mean age matched that of the working population but they were one of two groups with the highest scores in all four areas of worrying (see Table 3 [Tab. 3]). The unemployed worried significantly more about their job and their financial situation than all other groups (analysis of covariance with age as covariate; post hoc pairwise comparisons, Bonferroni adjusted, p<.05). With respect to worries about the family, the unemployed differed significantly from pensioners and non-working persons, and from the full-time employed and trainees, pupils, and students, but not from the other groups. They do not differ from pensioners, the part-time employed, trainees, and persons in military service or parental leave with respect to their worries about health.

Worries and income

Worries decreased with increasing household income (see Table 4 [Tab. 4]). Here we see clear and significant but weak associations. Spearman correlation coefficients were -.01 (n.s.) for worrying about one’s job, -.17 for financial situation, -.09 for family, and -.16 for health (p<.001 for all three items).

The next statements about significant differences between groups were evaluated using analysis of covariance with age and number of persons belonging to the respondent’s household as covariates. The latter covariate was chosen because the total net income of all members of the household was asked about and not the income per capita. Post hoc comparisons were Bonferroni adjusted (p<.05).

Respondents in the group with the lowest income (less than 750 €/month) worried significantly more about their job, financial situation and health than all other groups. With respect to financial situation, the same applied for the group with incomes of between 750 and 1500 €/month. Both these groups do not differ from each other in terms of worries about their family but differed from all other groups.

Worries and gender

With exception of the financial worries women worried significantly (due to large number of cases) more than men about their family and health but less about their job. Absolute differences ranged from .13 to .18, with standard deviations of about .9 thus indicating small effect sizes. Therefore, we have refrained from reporting detailed statistics.

Living situation and worries

Table 5 [Tab. 5] shows the cross tabulation for living situation by income. About a third of all participants were living alone, and another third were living together with another adult in a two-person household. These proportions applied quite well to the full-time employed, but not to pensioners. Nearly all pensioners (98%) were living alone or together with another adult. Working part-time or not working was over represented in households with two adults and at least one child (>42% in these groups vs. 21.6% in the total sample).

The 75 single parents (respondents of age 18 and older living in households with one adult and at least on child) were the group who worried the most about their job, financial situation and family (see Table 6 [Tab. 6]). Worries about the family were significantly higher than those of every other group (analysis of covariance with age as covariate; post hoc comparisons are Bonferroni adjusted, p<.05). The single parents also worried significantly more about their financial situation than all other groups apart from singles living alone.

The least worries about job, financial situation and family were seen in members of two-person households without children. 736 of the 890 respondents (82.7%) from these households reported that both members earned money, so had two incomes at their disposal.

Regarding health, singles reported significantly higher grades of solicitudes than all other groups. However, 42% of the singles were pensioners, the group that worried most about health, as mentioned above. Adjusting for age, the singles differed significantly only from the two groups of households with two adults with and without children. The least concern about health was reported by members of households containing at least two adults with children.

Comparing households with children (n=694) to those without (n=1779) the first group reported more worries about their job (1.88 vs. 1.62), their financial situation (2.18 vs. 1.98) and their family (1.94 vs. 1.81). This relationship is the other way around for worries about health (1.72 vs. 2.03). Adjusting for age this picture changes somewhat (analysis of covariance with age as covariate; p<.05). Then, having children in the household has an effect only on worries about the family, but standard deviations about .9 indicate small effect sizes.

Education and worries

Adjusting for age (analysis of covariance with age as covariate; p<.05), persons with higher education (>12 years) worried less about their financial situation and their job than persons with less education. Absolute differences were .15 and .22 respectively, thus indicating small effects, therefore no more descriptive statistics are given.

Place of residence and worries

Previous research showed that persons living in former East Germany described their psychological state differently to those living in former West Germany (10-12). Respondents from East Germany worried more than those from West Germany concerning all four items. The difference was the smallest for worries about one’s job: 1.76 vs. 1.67, and largest for financial worries: 2.22 (SD=.97, n=567) vs. 1.98 (SD=.90, n=1889) indicating a weak effect only. The absolute differences were .18 for worrying about the family, and .11 for worries about one’s health (i.e. below weak effect sizes). Adjusting for age, all differences were significant (analysis of covariance with age as covariate; p<.05).

Worries: intercorrelations

All four items addressing worries were intercorrelated. As seen in Table 7 [Tab. 7], there were clear correlations for financial worries with both job and family worries.

Worries and chronic stress

Worries can be considered an expression of psychological distress. Therefore, it could be expected that the answers to the four items addressing worries would be positively correlated with the screening scale for chronic stress (TICS). These correlations showed up as moderate (.24<r<.33, n=2372, p<.001). A stepwise linear regression of the screening scale on the four items resulted in a model comprising worries about one’s job, family and health. This model accounted for 15% of the variance (multiple R=.39) and did not improve when financial worries were entered.

Worries and physical well-being

Our four items correlated differently with the FEW total score. This correlation was not significant for worries about one’s job (-0.02); but was -0.17 for financial worries, and -0.25 for worries about the family. The highest coefficient was with worrying about one’s health (-0.52). This was the only variable that was included in a stepwise regression model accounting for about 27% of the variance of the FEW total score.


Four questions used in our clinical practise and addressing worries about one’s job, financial situation, family, and health situation were given to a sample representative of the German general population. We are aware that different social subgroups may interpret these items in different manners, and that these worries may be of different relevance to each of them. Therefore we examined such subgroups and compared them with each other.

The older the respondent the more worries about health was reported. This reflects the fact that morbidity increases with age. Accordingly, pensioners reported the most worries about their health. Compared to these worries, financial worries were quite low. Both pensioners and non-working persons reported virtually no worries about their job, a fact that is not surprising considering that they need not to work for money (any more).

In general, worries were most prominent in the unemployed. Only 24 of the 168 unemployed persons (14.3%) reported worrying “not at all” or “a little bit” for all four items, compared to about 50% of the total sample. They clearly worry most about material issues (see Table 3 [Tab. 3]), i.e. financial situation and job.

The IAB (Institut für Arbeitsmarkt- und Berufsforschung der Bundesanstalt für Arbeit) reports that about every third unemployed person suffers from health problems and that these restrictions diminish chances of getting a job [10]. It is surprising that in the unemployed, the correlation between worrying about one’s job and one’s health is weak (r=0.11, not significant). This correlation is at a similar level to that of the total sample (r=0.10). In the full-time employed, this correlation is more pronounced (r=0.37) and reflects the (justified) worry that health problems may threaten one’s work-place. A study with patients of general practitioners in Saxony indicated the same results as well. In this study the employed worried more about their work-place than unemployed patients [11].

We were surprised by the fact that persons in military service or on parental leave (n=37) seem to show the same level of distress as the unemployed group. However, regarding the small sample size, this finding may have been due to chance.

The saying “money isn't everything” is commonly used. Happiness is not equivalent to the absence of worries, but our data indicate that the level of worry decreases as income increases. Thus, higher income may facilitate happiness. It could be speculated that higher income would increase worries because the size of the potential loss is also higher. However, our results (see Table 4 [Tab. 4]) do not support this hypothesis.

Together with the unemployed, the single parents stood out from the other groups. They reported the highest level of worry among all groups with respect to job, financial situation and family. In strong contrast to them were the households with exactly two adults. In most cases these households involved so called “dinks” (double income, no kids). The gap between available incomes is obvious. 72% of the single parents reported incomes of 1500 Euro per month or less, but only 29.1% of the two-person adult households did so. This fact reflects the influence of material aspects on a person’s psychological situation.

As expected, worries correlated with chronic stress but to a lesser extent than anticipated. Correlations were more pronounced with physical well-being and worries about health. These correlations may be interpreted as evidence that the respondents did not simply lament “across the board”, but assessed each of the four domains of worrying quite separately. Regarding the evident correlations (see Table 7 [Tab. 7]) between worries about one’s job and one’s financial situation (r=0.57) and between worries about one’s financial situation and one’s family (r=0.52), we understand our results as indicators of validity for our four questions.

In all, these four questions addressing worries about one’s job, family, financial situation, and health may be seen as screening questions. They focus on relevant aspects of a person’s life situation that considerably influence one’s psychological state. A high level of worrying could impede their access to inner resources, thereby hindering the course of recovery. It seems plausible that a high degree of worry may be relevant with respect to disorders such as those classified in the ICD-10. However, this assumption has not been proven and needs further empirical investigation.

Our screening instrument is also of relevance to physical and psychological disorders. It offers a starting-point to support coping efforts. A routine screening would prevent real external and social stress in all-day clinical practise from being overlooked. Depending on the level of worry and in which domain, targeted support could be made available. For example, worries mainly about one’s financial situation or job could indicate that social services should be contacted, and predominant worries about the family may represent an indication for marital or family counselling.

Our study is limited by the fact that we did not know anything about the respondents’ individual concept of worries. We are aware that, for example, the question on worrying about one’s job has different meaning for the employed and than for unemployed persons, or even pensioners. The same applies to the question on worries about the family for 18 year-olds compared to 80 year-olds. The clear associations between worries and age would support this conjecture. If worrying “not at all” or “a little bit” is reported, no special action has to be taken in the clinical setting. We suggest further examination only if there are signs of high distress (“quite a lot” or “very much”). Our questions do not allow for a complex analysis of worries in special subgroups. Therefore more detailed questions and focusing on the special target groups are needed.


Stöber J. Besorgnis: Ein Vergleich dreier Inventare zur Erfassung allgemeiner Sorgen. Zeitschrift für differentielle und diagnostische Psychologie. 1995;16(1):50-63.
Liebert RM, Morris LW. Cognitive and emotional components of test anxiety: a distinction and some initial data. Psychol Rep. 1967;20:975-8.
Borcovec TD, Ray WJ, Stöber J. Worry: A cognitive phenomenon intimately linked to affective, physiological, and interpersonal behavioral processes. Cognit Ther Res. 1998;22(6):561-76.
Becker ES, Hoyer J. Generalisierte Angststörung. Göttingen: Hogrefe; 2005.
Schulz P, Schlotz W, Becker P. Trierer Inventar zum chronischen Streß (TICS). Göttingen, Bern, Toronto, Seattle: Hogrefe; 2004.
Kolip P, Schmidt B. Der Fragebogen zur Erfassung körperlichen Wohlbefindens. Zeitschrift für Gesundheitspsychologie. 1999;7:77-87.
Albani C, Blaser G, Geyer M, Schmutzer G, Hinz A, Bailer H, Grulke N, Brähler E. Validierung und Normierung des "Fragebogen zur Erfassung des körperlichen Wohlbefindens" (FEW-16) von Kolip und Schmidt an einer repräsentativen deutschen Bevölkerungsstichprobe. Psychother Psychosom Med Psychol. 2006;56(3-4):172-81.
ADM Arbeitskreis Deutscher Markt- und Sozialforschungsinstitute e. V. Jahresbericht 2003. Frankfurt: ADM; 2003. Available from:
Cohen J. Statistical power analysis for the behavioral sciences. 2 ed. Hillsdale, New Jersey: Lawrence Erlbaum Associates; 1988.
Hollederer A. Arbeitslos - Gesundheit los - chancenlos? IAB Kurzbericht. 2003(4):1-6.
Harych H. Sorge um den Arbeitsplatz, Arbeitslosigkeit und gesundheitliches Befinden - erste Ergebnisse einer Studie in Sachsen. Gesundheitswesen. 1995;57(2):82-5.
Schumacher J, Laubach W, Brähler E. Wie zufrieden sind wir mit unserem Leben? Soziodemographische und psychologische Prädiktoren der allgemeinen und bereichsspezifischen Lebenszufriedenheit. Z Med Psychol. 1995;4:17-26.
Gunzelmann T, Schumacher J, Strauß B, Brähler E. Interpersonale Probleme älterer Menschen. Psychotherapeut. 2000;45:153-64.
Albani C, Blaser G, Körner A, Geyer M, Volkart R, O'Connor L, Berry J, Brähler E. Der Fragebogen zu interpersonellen Schuldgefühlen (FIS). Anwendung in einer repräsentativen Bevölkerungsstichprobe und bei PsychotherapiepatientInnen. Psychother Psychosom Med Psychol. 2002;52:189-97.